{ "features_resolved": { "snapshot_any": true, "snapshot_mem": false, "fork": false, "pause_resume": true, "persistent_disk": true, "whole_disk": null, "persist_scope": null, "volumes": true, "auto_stop_idle": false, "long_sessions": true, "custom_image": true, "image_snapshot": true, "image_oci": true, "vm_isolation": false, "docker_inside": false, "nested_virt": false, "root": null, "systemd": null, "browser": true, "desktop": true, "computer_use": false, "code_interpreter": null, "browser_control": null, "desktop_control": null, "browser_and_desktop": false, "anti_bot": false, "captcha_solving": false, "residential_ip": true, "agent_harnesses": false, "ssh": true, "public_ipv4": false, "inbound_https": true, "custom_domain": null, "raw_tcp_inbound": null, "egress_allowlist": false, "open_internet": true, "static_egress_ip": true, "private_network": null, "self_host": false, "byoc": false, "open_source": false, "soc2": true, "hipaa": true, "sso": false, "gpu": true, "arm64": true, "windows": false, "macos": false, "lang_python": true, "lang_node": false, "lang_go": null, "lang_rust": null, "lang_java": null, "gpu_desktop": null, "credential_injection": null, "egress_http_rules": false, "wake_on_request": false, "live_resize": false, "fork_running": false, "memory_snapshot_fork": false, "webhooks": null, "scheduled_wakeups": null, "audit_logs": null, "eu_data_residency": true, "zero_data_retention": null, "scoped_api_keys": null, "spend_limits": null, "mcp_server": false, "usage_api": true, "snapshot_auto": false, "snapshot_on_demand": true, "harness_api": false, "own_agent_api": false, "fast_boot": null, "strong_isolation": false, "ingress_rules": true, "guest_firewall": true, "secret_proxy": false, "secret_proxy_any": false, "volume_attach": true, "volume_shared": false, "sidecar": null, "fuse": null, "kernel_features": null, "burst": null }, "company": { "id": "vast-ai", "checked": "2026-09-30", "company": "Vast.ai, Inc.", "ownership": { "value": "independent", "parent": null, "source": "https://vast.ai/about", "quote": "On June 28, 2016, they incorporated Vast.ai. ... Jake Cannell and Christian Horne incorporate Vast.ai as a Delaware C Corporation." }, "stage": { "value": "bootstrapped", "source": "https://vast.ai/about", "quote": "Vast.ai launches — not with a press release, but the way honest products launch: to friends, family, and a post on Reddit. ... 2025 40+ employees, two offices, SOC 2 certified ... The infrastructure that makes AI open is now an institution.", "confidence": "low", "note": "Vast.ai's own About/press pages narrate a self-funded, founder-built growth story (2016 incorporation to 40+ employees / two offices by 2025) with no mention of any investor or funding round anywhere on the site, and the company runs a generic 'Investor Inquiries' contact form rather than announcing any raise. A Google AI-generated search summary asserted an unsourced ~$4M seed round in July 2024, but no primary source, press article, or aggregator page (Crunchbase/PitchBook were inaccessible; no TechCrunch/press coverage found) could be located to confirm this, so it is not treated as confirmed. Do not confuse with 'VAST Data' (enterprise storage, ~$1.4B raised, $30B valuation) or 'Vast Space' (aerospace) or the Chinese 3D-AI startup 'Tripo/Vast' — all unrelated companies with similar names." }, "accelerators": [], "total_funding_usd": { "value": null, "status": "unknown" }, "last_round": { "type": null, "amount_usd": null, "date": null, "leads": [], "status": "unknown", "source": "", "quote": "" }, "investors": [], "portfolio_listings": [], "revenue": { "value": null, "status": "unknown" }, "employees": { "value": "40+", "source": "https://vast.ai/about" }, "note": "GPU rental/cloud marketplace (peer-to-peer + data-center 'Secure Cloud'), founded 2016 by Jake Cannell (CEO) and Christian Horne, launched publicly Sept 2018; Travis Cannell (COO) joined 2022. HQ Los Angeles (since 2024) plus a San Francisco engineering office (2025); SOC 2 Type I certified (2025). No accelerator affiliation, funding round, or named investor could be confirmed from primary sources within the research window; company presents itself as an organically grown, founder-built company. Search-engine access in this environment was heavily rate-limited/blocked (Google, Bing, DuckDuckGo, Crunchbase, PitchBook, Tracxn largely inaccessible), which constrained verification — treat funding/revenue fields as an incomplete picture rather than proof of 'no funding ever raised'." }, "id": "vast-ai", "name": "Vast.ai", "url": "https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx", "category": "gpu-cloud", "isolation": "container", "modes": [ { "key": "on-demand-rtx4090-0", "label": "RTX4090, on-demand", "flags": [ "alt" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "on-demand-rtx5090-1", "label": "RTX5090, on-demand", "flags": [ "alt" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "on-demand-rtx-pro-6000-2", "label": "RTX-PRO-6000, on-demand", "flags": [ "alt" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "on-demand-a100-80gb-3", "label": "A100-80GB, on-demand", "flags": [ "alt" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "on-demand-h100-4", "label": "H100, on-demand", "flags": [ "alt" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "on-demand-h200-5", "label": "H200, on-demand", "flags": [ "alt" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "on-demand-b200-6", "label": "B200, on-demand", "flags": [ "alt" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "spot-rtx4090-7", "label": "RTX4090, spot", "flags": [ "alt", "spot" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "spot-rtx5090-8", "label": "RTX5090, spot", "flags": [ "alt", "spot" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "spot-rtx-pro-6000-9", "label": "RTX-PRO-6000, spot", "flags": [ "alt", "spot" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "spot-a100-80gb-10", "label": "A100-80GB, spot", "flags": [ "alt", "spot" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "spot-h100-11", "label": "H100, spot", "flags": [ "alt", "spot" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "spot-h200-12", "label": "H200, spot", "flags": [ "alt", "spot" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "spot-b200-13", "label": "B200, spot", "flags": [ "alt", "spot" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "reserved-rtx4090-14", "label": "RTX4090, reserved", "flags": [ "alt", "sales" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": true, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "reserved-rtx5090-15", "label": "RTX5090, reserved", "flags": [ "alt", "sales" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": true, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "reserved-rtx-pro-6000-16", "label": "RTX-PRO-6000, reserved", "flags": [ "alt", "sales" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": true, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "reserved-a100-80gb-17", "label": "A100-80GB, reserved", "flags": [ "alt", "sales" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": true, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "reserved-h100-18", "label": "H100, reserved", "flags": [ "alt", "sales" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": true, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "reserved-h200-19", "label": "H200, reserved", "flags": [ "alt", "sales" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": true, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "reserved-b200-20", "label": "B200, reserved", "flags": [ "alt", "sales" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": true, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "serverless-rtx4090-21", "label": "RTX4090, serverless", "flags": [ "alt" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "serverless-rtx5090-22", "label": "RTX5090, serverless", "flags": [ "alt" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "serverless-rtx-pro-6000-23", "label": "RTX-PRO-6000, serverless", "flags": [ "alt" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "serverless-a100-80gb-24", "label": "A100-80GB, serverless", "flags": [ "alt" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "serverless-h100-25", "label": "H100, serverless", "flags": [ "alt" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "serverless-h200-26", "label": "H200, serverless", "flags": [ "alt" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "serverless-b200-27", "label": "B200, serverless", "flags": [ "alt" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": null, "ram_gib_h": null, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": null, "max_ram_gib": null, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": null, "note": "Fetch a live offer with location, reliability, count, CPU/RAM, storage and bandwidth rates; null avoids inventing an evergreen market price. Source: https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx ESTIMATOR EXCLUDED: Not a fixed verified USD quote. Bundled node count/CPU/RAM constraints are not enforceable in current engine schema. CPU/RAM inclusion not verified. Exact researched rate remains in gpu.json." }, { "key": "on-demand-h100", "label": "H100 SXM, on-demand marketplace (x1, from $3.09/h)", "flags": [ "stock" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": 0, "ram_gib_h": 0, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": 24, "max_ram_gib": 125.94, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": { "H100": 3.133333 }, "note": "Cheapest verified (reliability>=0.98) rentable 1x H100 SXM on-demand offer on 2026-09-28: dph_base $3.0907/h (compute incl. its host CPU/RAM share; dph_total with default storage $3.0980), 24 effective vCPU / 125.9 GiB, host storage $0.664/GB-month, upload $0.0378/GB. Collapses the former two-offer snapshot regimes (identical GPU, same or near-same price). No uptime/stock guarantee. Source: https://console.vast.ai/api/v0/bundles/?q=%7B%22verified%22%3A%7B%22eq%22%3Atrue%7D%2C%22rentable%22%3A%7B%22eq%22%3Atrue%7D%2C%22rented%22%3A%7B%22eq%22%3Afalse%7D%2C%22reliability%22%3A%7B%22gte%22%3A0.98%7D%2C%22num_gpus%22%3A%7B%22eq%22%3A1%7D%2C%22gpu_name%22%3A%7B%22eq%22%3A%22H100%20SXM%22%7D%2C%22type%22%3A%22on-demand%22%2C%22order%22%3A%5B%5B%22dph_total%22%2C%22asc%22%5D%5D%2C%22limit%22%3A20%7D", "stock_note": "marketplace offer snapshot 2026-09-28 (4 verified 1-GPU offers, cheapest used); offers come and go" }, { "key": "on-demand-h200", "label": "H200, on-demand marketplace (x1, from $4.80/h)", "flags": [ "stock" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": 0, "ram_gib_h": 0, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": 28, "max_ram_gib": 377.93, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": { "H200": 4.276316 }, "note": "Cheapest verified (reliability>=0.98) rentable 1x H200 on-demand offer on 2026-09-28: dph_base $4.8026/h (compute incl. its host CPU/RAM share; dph_total with default storage $4.8056), 28 effective vCPU / 377.9 GiB, host storage $0.263/GB-month, upload $0.0039/GB. Collapses the former two-offer snapshot regimes (identical GPU, same or near-same price). No uptime/stock guarantee. Source: https://console.vast.ai/api/v0/bundles/?q=%7B%22verified%22%3A%7B%22eq%22%3Atrue%7D%2C%22rentable%22%3A%7B%22eq%22%3Atrue%7D%2C%22rented%22%3A%7B%22eq%22%3Afalse%7D%2C%22reliability%22%3A%7B%22gte%22%3A0.98%7D%2C%22num_gpus%22%3A%7B%22eq%22%3A1%7D%2C%22gpu_name%22%3A%7B%22eq%22%3A%22H200%22%7D%2C%22type%22%3A%22on-demand%22%2C%22order%22%3A%5B%5B%22dph_total%22%2C%22asc%22%5D%5D%2C%22limit%22%3A20%7D", "stock_note": "marketplace offer snapshot 2026-09-28 (2 verified 1-GPU offers, cheapest used); offers come and go" }, { "key": "on-demand-b200", "label": "B200, on-demand marketplace (x1, from $6.25/h)", "flags": [ "stock" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": 0, "ram_gib_h": 0, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": 24, "max_ram_gib": 249.5, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": { "B200": 7.5 }, "note": "Cheapest verified (reliability>=0.98) rentable 1x B200 on-demand offer on 2026-09-28: dph_base $6.2500/h (compute incl. its host CPU/RAM share; dph_total with default storage $6.2521), 24 effective vCPU / 249.5 GiB, host storage $0.187/GB-month, upload $0.0037/GB. Collapses the former two-offer snapshot regimes (identical GPU, same or near-same price). No uptime/stock guarantee. Source: https://console.vast.ai/api/v0/bundles/?q=%7B%22verified%22%3A%7B%22eq%22%3Atrue%7D%2C%22rentable%22%3A%7B%22eq%22%3Atrue%7D%2C%22rented%22%3A%7B%22eq%22%3Afalse%7D%2C%22reliability%22%3A%7B%22gte%22%3A0.98%7D%2C%22num_gpus%22%3A%7B%22eq%22%3A1%7D%2C%22gpu_name%22%3A%7B%22eq%22%3A%22B200%22%7D%2C%22type%22%3A%22on-demand%22%2C%22order%22%3A%5B%5B%22dph_total%22%2C%22asc%22%5D%5D%2C%22limit%22%3A20%7D", "stock_note": "marketplace offer snapshot 2026-09-28 (3 verified 1-GPU offers, cheapest used); offers come and go" }, { "key": "on-demand-a100-sxm4", "label": "A100 SXM4, on-demand marketplace (x1, from $0.43/h)", "flags": [ "stock" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": 0, "ram_gib_h": 0, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": 24, "max_ram_gib": 62.96, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": { "A100-40GB": 0.426667 }, "note": "Cheapest verified (reliability>=0.98) rentable 1x A100 SXM4 on-demand offer on 2026-09-28: dph_base $0.4267/h (compute incl. its host CPU/RAM share; dph_total with default storage $0.4296), 24 effective vCPU / 63.0 GiB, host storage $0.267/GB-month, upload $0.0065/GB. Collapses the former two-offer snapshot regimes (identical GPU, same or near-same price). No uptime/stock guarantee. Source: https://console.vast.ai/api/v0/bundles/?q=%7B%22verified%22%3A%7B%22eq%22%3Atrue%7D%2C%22rentable%22%3A%7B%22eq%22%3Atrue%7D%2C%22rented%22%3A%7B%22eq%22%3Afalse%7D%2C%22reliability%22%3A%7B%22gte%22%3A0.98%7D%2C%22num_gpus%22%3A%7B%22eq%22%3A1%7D%2C%22gpu_name%22%3A%7B%22eq%22%3A%22A100%20SXM4%22%7D%2C%22type%22%3A%22on-demand%22%2C%22order%22%3A%5B%5B%22dph_total%22%2C%22asc%22%5D%5D%2C%22limit%22%3A20%7D", "stock_note": "marketplace offer snapshot 2026-09-28 (11 verified 1-GPU offers, cheapest used); offers come and go" }, { "key": "on-demand-rtx4090", "label": "RTX 4090, on-demand marketplace (x1, from $0.39/h)", "flags": [ "stock" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": 0, "ram_gib_h": 0, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": 12, "max_ram_gib": 62.89, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": { "RTX-4090": 0.346667 }, "note": "Cheapest verified (reliability>=0.98) rentable 1x RTX 4090 on-demand offer on 2026-09-28: dph_base $0.3867/h (compute incl. its host CPU/RAM share; dph_total with default storage $0.3941), 12 effective vCPU / 62.9 GiB, host storage $0.667/GB-month, upload $0.0026/GB. Collapses the former two-offer snapshot regimes (identical GPU, same or near-same price). No uptime/stock guarantee. Source: https://console.vast.ai/api/v0/bundles/?q=%7B%22verified%22%3A%7B%22eq%22%3Atrue%7D%2C%22rentable%22%3A%7B%22eq%22%3Atrue%7D%2C%22rented%22%3A%7B%22eq%22%3Afalse%7D%2C%22reliability%22%3A%7B%22gte%22%3A0.98%7D%2C%22num_gpus%22%3A%7B%22eq%22%3A1%7D%2C%22gpu_name%22%3A%7B%22eq%22%3A%22RTX%204090%22%7D%2C%22type%22%3A%22on-demand%22%2C%22order%22%3A%5B%5B%22dph_total%22%2C%22asc%22%5D%5D%2C%22limit%22%3A20%7D", "stock_note": "marketplace offer snapshot 2026-09-28 (20 verified 1-GPU offers, cheapest used); offers come and go" }, { "key": "on-demand-rtx5090", "label": "RTX 5090, on-demand marketplace (x1, from $0.47/h)", "flags": [ "stock" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": 0, "ram_gib_h": 0, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": 30, "max_ram_gib": 94.22, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": { "RTX-5090": 0.44 }, "note": "Cheapest verified (reliability>=0.98) rentable 1x RTX 5090 on-demand offer on 2026-09-28: dph_base $0.4667/h (compute incl. its host CPU/RAM share; dph_total with default storage $0.4686), 30 effective vCPU / 94.2 GiB, host storage $0.173/GB-month, upload $0.0130/GB. Collapses the former two-offer snapshot regimes (identical GPU, same or near-same price). No uptime/stock guarantee. Source: https://console.vast.ai/api/v0/bundles/?q=%7B%22verified%22%3A%7B%22eq%22%3Atrue%7D%2C%22rentable%22%3A%7B%22eq%22%3Atrue%7D%2C%22rented%22%3A%7B%22eq%22%3Afalse%7D%2C%22reliability%22%3A%7B%22gte%22%3A0.98%7D%2C%22num_gpus%22%3A%7B%22eq%22%3A1%7D%2C%22gpu_name%22%3A%7B%22eq%22%3A%22RTX%205090%22%7D%2C%22type%22%3A%22on-demand%22%2C%22order%22%3A%5B%5B%22dph_total%22%2C%22asc%22%5D%5D%2C%22limit%22%3A20%7D", "stock_note": "marketplace offer snapshot 2026-09-28 (20 verified 1-GPU offers, cheapest used); offers come and go" }, { "key": "on-demand-rtx-pro-6000", "label": "RTX PRO 6000 S, on-demand marketplace (x1, from $1.40/h)", "flags": [ "stock" ], "os": [ "linux" ], "regions": null, "gpu_only": true, "features": {}, "pricing": "resource", "vcpu_h": 0, "ram_gib_h": 0, "cpu_basis": "alloc", "ram_basis": "alloc", "active_floor": 0, "ram_per_vcpu": [ null, null ], "ram_fixed_per_vcpu": null, "vcpu_options": null, "min_vcpu": null, "max_vcpu": 16, "max_ram_gib": 98.34, "sizes": null, "pool_tiers": null, "requires_always_on": false, "always_on_month_per_instance": null, "commit_note": null, "min_billed_seconds": null, "granularity_s": 1, "start_fee": null, "boot_overhead_s": null, "multiplier": 1, "windows_vcpu_h": null, "gpu": { "RTX-PRO-6000": 1.4 }, "note": "Cheapest verified (reliability>=0.98) rentable 1x RTX PRO 6000 S on-demand offer on 2026-09-28: dph_base $1.4000/h (compute incl. its host CPU/RAM share; dph_total with default storage $1.4022), 16 effective vCPU / 98.3 GiB, host storage $0.200/GB-month, upload $0.0039/GB. Collapses the former two-offer snapshot regimes (identical GPU, same or near-same price). No uptime/stock guarantee. Source: https://console.vast.ai/api/v0/bundles/?q=%7B%22verified%22%3A%7B%22eq%22%3Atrue%7D%2C%22rentable%22%3A%7B%22eq%22%3Atrue%7D%2C%22rented%22%3A%7B%22eq%22%3Afalse%7D%2C%22reliability%22%3A%7B%22gte%22%3A0.98%7D%2C%22num_gpus%22%3A%7B%22eq%22%3A1%7D%2C%22gpu_name%22%3A%7B%22eq%22%3A%22RTX%20PRO%206000%20S%22%7D%2C%22type%22%3A%22on-demand%22%2C%22order%22%3A%5B%5B%22dph_total%22%2C%22asc%22%5D%5D%2C%22limit%22%3A20%7D", "stock_note": "marketplace offer snapshot 2026-09-28 (11 verified 1-GPU offers, cheapest used); offers come and go" } ], "storage": { "included_disk_gib": 0, "disk_gib_month": 0.3167, "disk_billed_when_stopped": true, "snapshot_gib_month": null, "paused_ram_billed": null }, "network": { "egress_free_gib": null, "egress_gib": 0.0052, "ipv4_month": null, "ipv4_available": null, "internet": { "default": "full" } }, "plans": [ { "name": "Pay as you go", "fee": 0, "flags": [], "fee_is_credit": false, "included_usd": 0, "per_seat": 0, "concurrency": null, "max_session_h": null, "max_vcpu": null, "max_ram_gib": null, "sizes_allowed": null, "note": "Self-serve pay-as-you-go account, no platform fee; GPU quota may need a request." } ], "free": { "monthly_credit": null, "one_time_credit": null }, "features": { "linux": true, "windows": false, "macos": false, "arm64": true, "gpu": true, "gpu_types": [ "A100-40GB", "A100-80GB", "B200", "H100", "H200", "RTX-PRO-6000", "RTX4090", "RTX5090" ], "snapshot": "fs", "pause_resume": true, "fork": false, "persistent_disk": true, "volumes": null, "auto_stop_idle": false, "max_session_h": null, "custom_image": true, "docker_inside": false, "nested_virt": false, "root": null, "systemd": null, "browser": true, "desktop": true, "computer_use": false, "code_interpreter": null, "agent_harnesses": false, "ssh": true, "public_ipv4": false, "inbound_https": true, "custom_domain": null, "raw_tcp_inbound": null, "egress_allowlist": false, "static_egress_ip": true, "private_network": null, "self_host": false, "byoc": false, "open_source": false, "regions": [ "other", "eu", "asia", "us" ], "soc2": true, "hipaa": true, "sso": false, "lang_python": true, "lang_node": false, "lang_go": null, "lang_rust": null, "lang_java": null, "gpu_desktop": null, "credential_injection": null, "egress_http_rules": false, "wake_on_request": false, "live_resize": false, "fork_running": false, "memory_snapshot_fork": false, "webhooks": null, "scheduled_wakeups": null, "audit_logs": null, "eu_data_residency": true, "zero_data_retention": null, "scoped_api_keys": null, "spend_limits": null, "mcp_server": false, "snapshot_auto": false, "snapshot_on_demand": true, "harness_api": false, "ingress_rules": true, "secret_proxy": false, "volume_attach": true, "isolation": "container", "image_oci": true, "anti_bot": false, "captcha_solving": false, "egress_ip_type": "residential+datacenter", "volume_shared": false, "own_agent_api": false, "guest_firewall": true, "max_session_unlimited": true, "open_internet": true, "usage_api": true }, "perf": { "cpu_runs_s": null, "cold_start_ms": null, "burst_tti_ms": null, "dax_s": null }, "knobs": [], "caveats": [ "No universal list rate: GPU, CPU/RAM share, bandwidth and storage are host-offer-specific. Per-second compute; storage continues while stopped. Serverless uses underlying marketplace workers with no extra surcharge. Reserved prepayment up to 50% off is not a fixed tariff. Dated public API snapshots added for 1 GPU, verified hosts, reliability>=0.98; first two cheapest offers by total listed hourly price. dph_base is compute, dph_total adds default 10GB storage; bandwidth remains extra. gpu_frac describes host share, NOT fractional physical GPU.", "Any null count, CPU/RAM, billing increment or minimum is unverified, not unlimited/free.", "This is GPU-product research, not a claim that every feature of the entire vendor documentation was audited. Unverified features are null.", "GPU multi-count bundles, variant/region selection and fractional slices require a SKU-aware estimator; unsupported rows remain unpriced in strict cards.", "Modes with gpu=null are intentionally excluded, not free GPU modes. Never substitute 0 for a null rate. Flags alt indicate DIY/ML infrastructure, not the same thing as managed untrusted-code sandboxes.", "Unknown fee, storage or bandwidth means full monthly total may be unavailable. Rates exclude taxes.", "2026-09-28 fix pass: snapshot regimes collapsed to one per GPU (cheapest live verified offer, flag stock). Disk priced at $0.2667/GB-month = median host storage_cost across 71 live verified 1-GPU offers (range $0.08-$1.00/GB-month; the host you rent sets it; storage keeps billing while stopped). Egress $0.0052/GB = median host inet_up_cost across the same offers (host-specific; download also billed per host). Live prices moved since the earlier snapshot (e.g. H100 SXM $1.87 -> $3.09/h, H200 $3.55 -> $4.80/h)." ], "sources": [ "https://github.com/vast-ai/docs/blob/main/guides/pricing.mdx", "https://docs.vast.ai/guides/pricing", "https://docs.vast.ai/llms.txt", "https://console.vast.ai/api/v0/bundles/?q=%7B%22verified%22%3A%7B%22eq%22%3Atrue%7D%2C%22rentable%22%3A%7B%22eq%22%3Atrue%7D%2C%22rented%22%3A%7B%22eq%22%3Afalse%7D%2C%22reliability%22%3A%7B%22gte%22%3A0.98%7D%2C%22num_gpus%22%3A%7B%22eq%22%3A1%7D%2C%22gpu_name%22%3A%7B%22eq%22%3A%22H100%20SXM%22%7D%2C%22type%22%3A%22on-demand%22%2C%22order%22%3A%5B%5B%22dph_total%22%2C%22asc%22%5D%5D%2C%22limit%22%3A20%7D", "https://console.vast.ai/api/v0/bundles/?q=%7B%22verified%22%3A%7B%22eq%22%3Atrue%7D%2C%22rentable%22%3A%7B%22eq%22%3Atrue%7D%2C%22rented%22%3A%7B%22eq%22%3Afalse%7D%2C%22reliability%22%3A%7B%22gte%22%3A0.98%7D%2C%22num_gpus%22%3A%7B%22eq%22%3A1%7D%2C%22gpu_name%22%3A%7B%22eq%22%3A%22H200%22%7D%2C%22type%22%3A%22on-demand%22%2C%22order%22%3A%5B%5B%22dph_total%22%2C%22asc%22%5D%5D%2C%22limit%22%3A20%7D", "https://console.vast.ai/api/v0/bundles/?q=%7B%22verified%22%3A%7B%22eq%22%3Atrue%7D%2C%22rentable%22%3A%7B%22eq%22%3Atrue%7D%2C%22rented%22%3A%7B%22eq%22%3Afalse%7D%2C%22reliability%22%3A%7B%22gte%22%3A0.98%7D%2C%22num_gpus%22%3A%7B%22eq%22%3A1%7D%2C%22gpu_name%22%3A%7B%22eq%22%3A%22B200%22%7D%2C%22type%22%3A%22on-demand%22%2C%22order%22%3A%5B%5B%22dph_total%22%2C%22asc%22%5D%5D%2C%22limit%22%3A20%7D", "https://console.vast.ai/api/v0/bundles/?q=%7B%22verified%22%3A%7B%22eq%22%3Atrue%7D%2C%22rentable%22%3A%7B%22eq%22%3Atrue%7D%2C%22rented%22%3A%7B%22eq%22%3Afalse%7D%2C%22reliability%22%3A%7B%22gte%22%3A0.98%7D%2C%22num_gpus%22%3A%7B%22eq%22%3A1%7D%2C%22gpu_name%22%3A%7B%22eq%22%3A%22A100%20SXM4%22%7D%2C%22type%22%3A%22on-demand%22%2C%22order%22%3A%5B%5B%22dph_total%22%2C%22asc%22%5D%5D%2C%22limit%22%3A20%7D", "https://console.vast.ai/api/v0/bundles/?q=%7B%22verified%22%3A%7B%22eq%22%3Atrue%7D%2C%22rentable%22%3A%7B%22eq%22%3Atrue%7D%2C%22rented%22%3A%7B%22eq%22%3Afalse%7D%2C%22reliability%22%3A%7B%22gte%22%3A0.98%7D%2C%22num_gpus%22%3A%7B%22eq%22%3A1%7D%2C%22gpu_name%22%3A%7B%22eq%22%3A%22RTX%204090%22%7D%2C%22type%22%3A%22on-demand%22%2C%22order%22%3A%5B%5B%22dph_total%22%2C%22asc%22%5D%5D%2C%22limit%22%3A20%7D", "https://console.vast.ai/api/v0/bundles/?q=%7B%22verified%22%3A%7B%22eq%22%3Atrue%7D%2C%22rentable%22%3A%7B%22eq%22%3Atrue%7D%2C%22rented%22%3A%7B%22eq%22%3Afalse%7D%2C%22reliability%22%3A%7B%22gte%22%3A0.98%7D%2C%22num_gpus%22%3A%7B%22eq%22%3A1%7D%2C%22gpu_name%22%3A%7B%22eq%22%3A%22RTX%205090%22%7D%2C%22type%22%3A%22on-demand%22%2C%22order%22%3A%5B%5B%22dph_total%22%2C%22asc%22%5D%5D%2C%22limit%22%3A20%7D", "https://console.vast.ai/api/v0/bundles/?q=%7B%22verified%22%3A%7B%22eq%22%3Atrue%7D%2C%22rentable%22%3A%7B%22eq%22%3Atrue%7D%2C%22rented%22%3A%7B%22eq%22%3Afalse%7D%2C%22reliability%22%3A%7B%22gte%22%3A0.98%7D%2C%22num_gpus%22%3A%7B%22eq%22%3A1%7D%2C%22gpu_name%22%3A%7B%22eq%22%3A%22RTX%20PRO%206000%20S%22%7D%2C%22type%22%3A%22on-demand%22%2C%22order%22%3A%5B%5B%22dph_total%22%2C%22asc%22%5D%5D%2C%22limit%22%3A20%7D" ], "fleet": { "default_max_instances": null, "default_max_vcpu": null, "default_max_gpus": null, "raise": "automatic", "raise_time_days": null, "verification": "Verify email and prepay credits before renting.", "tiers": null, "notes": "New accounts have a spend-rate limit, especially low before email verification; it grows over time. Docs suggest cheaper instance or waiting hours, then support chat if needed. No numeric default instance/GPU cap or guaranteed quota-raise time published.", "sources": [ "https://github.com/vast-ai/docs/blob/main/guides/reference/troubleshooting.mdx", "https://github.com/vast-ai/docs/blob/main/guides/instances/pricing.mdx" ], "researched": "2026-09-28" }, "audit_2026_09_29": { "applied": 5, "source": "checked against the provider's live pricing and docs pages", "reviewed": 1 }, "feature_evidence": { "snapshot_auto": { "value": false, "source": "https://docs.vast.ai/guides/instances/manage-instances", "quote": "Stop Button (square icon): Pauses instance, preserves data, continues storage charges", "note": "Stopping keeps the container disk on the host (persistent_disk), processes/memory are not saved; no automatic checkpoint on stop/idle is documented. No idle auto-stop.", "checked": "2026-09-29" }, "snapshot_on_demand": { "value": true, "source": "https://github.com/vast-ai/vast-cli/blob/master/vast.py", "quote": "Takes a snapshot of a running container instance and pushes snapshot to the specified repository in container registry.", "note": "vastai take snapshot INSTANCE_ID --repo ... (POST /instances/take_snapshot/{id}/): docker-commit style, filesystem only (optionally pauses processes during commit), pushed to YOUR registry (docker.io default); new instances can boot from that image. Found in the official CLI source, not in the docs site (the docs page for take-snapshot 404s). Also volume clone (vastai clone volume) and vastai copy for data.", "checked": "2026-09-29" }, "harness_api": { "value": false, "source": "https://docs.vast.ai/guides/get-started/agents", "quote": "A **skill** or **plugin** lets an agent operate your **account** — find, rent, launch, and tear down instances.", "note": "Vast's agent integration is the reverse direction (skills/plugins for Claude Code/Codex/Cursor that drive the Vast CLI). Inside instances only SSH/Jupyter and a constrained 'vastai execute' command API; no API to drive an agent harness in the instance.", "checked": "2026-09-29" }, "egress_allowlist": { "value": false, "source": "https://docs.vast.ai/guides/instances/connect/networking", "quote": "Vast.ai docker instances have full internet access, but generally do not have unique IP addresses.", "note": "No provider-side egress controls documented; Security FAQ: 'Network restrictions depend on the host configuration. Some options: Use SSH key authentication (no passwords); Configure firewall rules in your container; Select providers with static IPs for IP whitelisting'. Marketplace of third-party hosts.", "checked": "2026-09-29" }, "egress_http_rules": { "value": false, "source": "https://docs.vast.ai/guides/reference/faq/security", "quote": "Network restrictions depend on the host configuration.", "note": "No egress controls of any kind; no HTTP-level rules.", "checked": "2026-09-29" }, "ingress_rules": { "value": true, "source": "https://docs.vast.ai/guides/instances/connect/networking", "quote": "You can open custom ports for any docker image using -p arguments in the docker create/run options box in the image config editor pop-up menu.", "note": "Only control is choosing which ports are exposed (mapped to random external ports on a shared host IP, max 64). No provider-side IP allowlists or security groups (Security FAQ points to in-container firewall). Instance Portal tokens (OPEN_BUTTON_TOKEN) and Cloudflare quick tunnels are implemented inside Vast base images, not enforced by the platform.", "checked": "2026-09-29" }, "secret_proxy": { "value": false, "source": "https://docs.vast.ai/cli/reference/create-env-var", "quote": "Create a new user environment variable", "note": "Secrets are account-level environment variables injected into instances (and template env); no credential-injecting proxy. Security FAQ advises 'Don't store credentials in instances'.", "checked": "2026-09-29" }, "volume_attach": { "value": true, "source": "https://docs.vast.ai/guides/instances/storage/volumes", "quote": "We currently provide **local volumes only**, meaning: A volume is physically tied to the machine it was created on. It can only be attached to instances running on the same physical machine.", "note": "Local volumes survive instance destruction and can be reattached to new instances on the same host; fixed size; Docker instances only (not VMs); billed per GB while they exist at host-set rates. The official CLI also has 'vastai create network volume' / 'search network volumes' (network volume offers) not described in the docs guides; migration guide says 'Vast has no cross-host network volume.'", "checked": "2026-09-29" }, "isolation": { "value": "container", "source": "https://docs.vast.ai/guides/instances/docker-environment.md", "quote": "Vast.ai instances run as Linux Docker containers.", "note": "card already models the marketplace's Docker container instance product; separate opt-in VM mode (qemu/kvm) exists but is not the modeled product", "checked": "2026-09-30" }, "image_oci": { "value": true, "source": "https://docs.vast.ai/guides/reference/faq/technical.md", "quote": "Can I use my own Docker images? Yes! ... Use standard Docker Hub, GHCR, or other registries", "note": null, "checked": "2026-09-30" }, "docker_inside": { "value": false, "source": "https://docs.vast.ai/guides/reference/faq/technical.md", "quote": "Why can't I run Docker inside my instance? Docker-in-Docker is disabled for security.", "note": "the separate full-VM mode does support nested containerization, but that is not the modeled container-instance product", "checked": "2026-09-30" }, "nested_virt": { "value": false, "source": "https://docs.vast.ai/guides/reference/faq/security.md", "quote": "Clients are isolated in unprivileged Docker containers", "note": "derived: unprivileged Docker instances have no /dev/kvm; only the separate full-VM mode advertises nested containerization, not KVM-in-KVM", "checked": "2026-09-30" }, "desktop": { "value": true, "source": "https://docs.vast.ai/examples/guides/linux-virtual-desktop.md", "quote": "Browser Access (Recommended): Click the 'Open' button ... Choose between two browser-based viewers: Selkies WebRTC ... NoVNC ... VNC Client: Connect using any VNC client, Address: instance_ip:5900", "note": "available via the Ubuntu Desktop (VM) template", "checked": "2026-09-30" }, "browser": { "value": true, "source": "https://docs.vast.ai/examples/guides/linux-virtual-desktop.md", "quote": "Web Browsers: Firefox, Chrome", "note": "pre-installed in the Ubuntu Desktop (VM) template", "checked": "2026-09-30" }, "computer_use": { "value": false, "source": "https://docs.vast.ai/guides/instances/overview.md", "quote": "Instances are Docker containers that give you exclusive GPU access for training, inference, and development. Pay by the second, connect via SSH or Jupyter.", "note": "not offered: outside the product's scope; no screenshot/click desktop-driving API found in docs", "checked": "2026-09-30" }, "anti_bot": { "value": false, "source": "https://docs.vast.ai/guides/instances/overview.md", "quote": "Instances are Docker containers that give you exclusive GPU access for training, inference, and development.", "note": "not offered: outside the product's scope", "checked": "2026-09-30" }, "captcha_solving": { "value": false, "source": "https://docs.vast.ai/guides/instances/overview.md", "quote": "Instances are Docker containers that give you exclusive GPU access for training, inference, and development.", "note": "not offered: outside the product's scope", "checked": "2026-09-30" }, "egress_ip_type": { "value": "residential+datacenter", "source": "https://docs.vast.ai/guides/reference/faq/security.md", "quote": "Tier 4 datacenters have extensive physical and operational security. Individual hosts may have less formal security measures.", "note": "marketplace mixes vetted 'Secure Cloud' datacenter hosts and individual/community hosts (residential-type connections); no single guaranteed egress IP class", "checked": "2026-09-30" }, "ssh": { "value": true, "source": "https://docs.vast.ai/guides/instances/connect/ssh.md", "quote": "Learn how to securely connect to Vast.ai instances using SSH.", "note": null, "checked": "2026-09-30" }, "snapshot": { "value": "fs", "source": "https://docs.vast.ai/guides/instances/manage-instances.md", "quote": "Stop Button (square icon): Pauses instance, preserves data, continues storage charges", "note": "disk/filesystem is preserved automatically across stop/start; GPU and running processes are not kept alive (not a memory snapshot)", "checked": "2026-09-30" }, "fork": { "value": false, "source": "https://docs.vast.ai/guides/instances/storage/volumes.md", "quote": "We currently provide local volumes only, meaning: A volume is physically tied to the machine it was created on.", "note": "only a volume-clone CLI command exists (vastai clone volume); no whole-machine clone/fork feature found", "checked": "2026-09-30" }, "fork_running": { "value": false, "source": "https://docs.vast.ai/guides/instances/storage/volumes.md", "quote": "A volume is physically tied to the machine it was created on.", "note": "no running-machine clone feature found", "checked": "2026-09-30" }, "memory_snapshot_fork": { "value": false, "source": "https://docs.vast.ai/guides/instances/manage-instances.md", "quote": "Stop Button (square icon): Pauses instance, preserves data, continues storage charges", "note": "no memory-included clone/fork feature found", "checked": "2026-09-30" }, "pause_resume": { "value": true, "source": "https://docs.vast.ai/guides/instances/manage-instances.md", "quote": "Stop Button (square icon): Pauses instance, preserves data, continues storage charges", "note": "compute/GPU billing stops while stopped; storage billing continues", "checked": "2026-09-30" }, "wake_on_request": { "value": false, "source": "https://docs.vast.ai/guides/instances/manage-instances.md", "quote": "Inactive: Stopped but data preserved (can restart if GPU available)", "note": "stopped instances must be manually restarted; no auto-wake-on-request mechanism found", "checked": "2026-09-30" }, "volume_shared": { "value": false, "source": "https://docs.vast.ai/guides/instances/storage/volumes.md", "quote": "A volume is physically tied to the machine it was created on. It can only be attached to instances running on the same physical machine.", "note": "local, single-machine volumes; no documented simultaneous multi-instance shared mount", "checked": "2026-09-30" }, "auto_stop_idle": { "value": false, "source": "https://docs.vast.ai/guides/reference/faq/technical.md", "quote": "Destroy instances you no longer need to avoid being charged for idle resources.", "note": "no automatic idle-stop feature found; users are told to manually stop/destroy idle instances", "checked": "2026-09-30" }, "live_resize": { "value": false, "source": "https://docs.vast.ai/guides/instances/docker-environment.md", "quote": "Each instance has a disk storage allocation reserved at creation time. This allocation is static and cannot be changed after creation", "note": "CPU/RAM are also fixed proportional shares of the host tied to the GPU allocation, not independently resizable live", "checked": "2026-09-30" }, "agent_harnesses": { "value": false, "source": "https://docs.vast.ai/guides/get-started/agents.md", "quote": "Claude Code requires Node.js: curl -fsSL https://deb.nodesource.com/setup_22.x | bash - ; apt-get install -y nodejs ; npm install -g @anthropic-ai/claude-code", "note": "coding agents are documented as something you self-install on an instance, not a preinstalled template/offering", "checked": "2026-09-30" }, "own_agent_api": { "value": false, "source": "https://docs.vast.ai/guides/get-started/agents.md", "quote": "A skill or plugin lets an agent operate your account — find, rent, launch, and tear down instances.", "note": "the skills/CLI let a THIRD-PARTY coding agent control your Vast account; Vast does not offer its own browser/computer-use/research task agent", "checked": "2026-09-30" }, "mcp_server": { "value": false, "source": "https://docs.vast.ai/examples/mcp/dr-tulu.md", "quote": "DR-Tulu was trained end-to-end with an MCP server providing web search and page reading capabilities.", "note": "the only MCP server mentioned is a user-deployed example app running inside an instance; Vast does not ship its own first-party MCP server to control its sandboxes", "checked": "2026-09-30" }, "lang_python": { "value": true, "source": "https://github.com/vast-ai/base-image/", "quote": "Use Vast base images (also on DockerHub), which are pre-cached on many hosts", "note": "default/recommended templates (PyTorch, ComfyUI, vLLM, etc.) are Python-based and ship Python out of the box", "checked": "2026-09-30" }, "lang_node": { "value": false, "source": "https://docs.vast.ai/guides/get-started/agents.md", "quote": "Claude Code requires Node.js: curl -fsSL https://deb.nodesource.com/setup_22.x | bash - ; apt-get install -y nodejs", "note": "Node.js is not part of the standard ML base images and must be installed manually", "checked": "2026-09-30" }, "public_ipv4": { "value": false, "source": "https://docs.vast.ai/guides/instances/connect/networking.md", "quote": "Vast.ai docker instances have full internet access, but generally do not have unique IP addresses. ... each docker instance gets a fraction of a public IP address based on a subset of ports.", "note": null, "checked": "2026-09-30" }, "inbound_https": { "value": true, "source": "https://docs.vast.ai/guides/instances/connect/instance-portal.md", "quote": "a secure Cloudflare tunnel will be created for each of your instance's open ports ... formatted like this: https://four-randomly-selected-words.trycloudflare.com", "note": null, "checked": "2026-09-30" }, "guest_firewall": { "value": true, "source": "https://docs.vast.ai/guides/reference/faq/security.md", "quote": "Can I restrict network access to my instances? ... Configure firewall rules in your container", "note": null, "checked": "2026-09-30" }, "static_egress_ip": { "value": true, "source": "https://docs.vast.ai/guides/reference/faq/security.md", "quote": "Select providers with static IPs for IP whitelisting", "note": "host-dependent filterable attribute (static_ip), not guaranteed for every machine", "checked": "2026-09-30" }, "windows": { "value": false, "source": "https://docs.vast.ai/guides/instances/overview.md", "quote": "Instances are Docker containers that give you exclusive GPU access for training, inference, and development.", "note": "not offered: outside the product's scope; guest OS is Linux, the 'Windows SSH Guide' is only about the client OS used to connect", "checked": "2026-09-30" }, "macos": { "value": false, "source": "https://docs.vast.ai/guides/instances/overview.md", "quote": "Instances are Docker containers that give you exclusive GPU access for training, inference, and development.", "note": "not offered: outside the product's scope", "checked": "2026-09-30" }, "arm64": { "value": true, "source": "https://docs.vast.ai/llms-full.txt", "quote": "# arm64 cpu architecture only\nvastai search offers 'cpu_arch=arm64'", "note": null, "checked": "2026-09-30" }, "self_host": { "value": false, "source": "https://docs.vast.ai/guides/concepts.md", "quote": "The core vocabulary of the Vast.ai platform, marketplace, offers, instances, templates, endpoints, workers.", "note": "not offered: Vast is a hosted marketplace connecting renters to third-party hosts, not software you self-host", "checked": "2026-09-30" }, "byoc": { "value": false, "source": "https://docs.vast.ai/guides/concepts.md", "quote": "The core vocabulary of the Vast.ai platform, marketplace, offers, instances, templates, endpoints, workers.", "note": "not offered: Vast runs on marketplace hosts' own hardware, not deployed into a customer's cloud account", "checked": "2026-09-30" }, "open_source": { "value": false, "source": "https://github.com/vast-ai/vast-cli", "quote": "Use Vast base images (also on DockerHub), which are pre-cached on many hosts", "note": "the vastai CLI and base images are open source, but the core marketplace/scheduling platform is not; not self-hostable", "checked": "2026-09-30" }, "sso": { "value": false, "source": "https://docs.vast.ai/guides/reference/faq/security.md", "quote": "How do I secure my account? ... Use a strong, unique password ... Regularly rotate API keys", "note": "no SAML/OIDC single sign-on documented; account security relies on password/2FA/API keys", "checked": "2026-09-30" }, "eu_data_residency": { "value": true, "source": "https://docs.vast.ai/llms-full.txt", "quote": "geolocation | string | two letter country code; supports =, !=, in, notin", "note": "renters can filter/select hosts by country (including EU countries) via the geolocation offer filter", "checked": "2026-09-30" }, "max_session_h": { "value": "none", "source": "https://docs.vast.ai/guides/reference/faq/rental-types.md", "quote": "On-Demand (High Priority): Fixed price set by the host, Runs as long as you want, Cannot be interrupted", "note": "interruptible (spot) instances can be stopped by higher bids at any time, but on-demand/reserved instances have no fixed max duration", "checked": "2026-09-30" }, "soc2": { "value": true, "source": "https://vast.ai/compliance", "quote": "Vast.ai has completed SOC 2 Type 2 certification. This audit, conducted by an independent third party, verified that our security, availability, and confidentiality controls meet AICPA Trust Services Criteria over a sustained observation period.", "note": null, "checked": "2026-09-30" }, "hipaa": { "value": true, "source": "https://vast.ai/compliance", "quote": "Vast.ai supports HIPAA-covered workloads on our Secure Cloud tier. Technical safeguards — including data isolation, access controls, and audit logging — align with HIPAA requirements. Business Associate Agreements (BAAs) are available for qualifying customers.", "note": null, "checked": "2026-09-30" }, "usage_api": { "value": true, "source": "https://raw.githubusercontent.com/vast-ai/docs/main/api-reference/openapi.yaml", "quote": "Shows charges per instance, including GPU, storage, and bandwidth. Can filter by contract type (instance, volume, serverless) and date range. Returns per-instance charge results with cost breakdown.", "note": "granularity: per-sandbox; cost: yes; Vast AI provides /api/v0/charges endpoint that returns itemized charges per instance. Each instance has a unique contract ID. The endpoint returns GPU compute cost, storage cost, and bandwidth cost per individual sandbox/instance. Results can be filtered by contract type and date range via select_filters parameter.", "confidence": "high", "checked": "2026-10-01" } }, "locations": [ { "country": "AU", "state": null, "city": "Sydney", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "BE", "state": null, "city": "Brussels", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "CA", "state": "BC", "city": "Vancouver", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "CN", "state": null, "city": "Kunming", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "CZ", "state": null, "city": "Prague", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "ES", "state": null, "city": "Madrid", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "FR", "state": null, "city": "Paris", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "GB", "state": null, "city": "London", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "KR", "state": null, "city": "Seoul", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "NO", "state": null, "city": "Oslo", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "PL", "state": null, "city": "Warsaw", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "TW", "state": null, "city": "Taipei", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "US", "state": "AZ", "city": "Phoenix", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "US", "state": "FL", "city": "Miami", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "US", "state": "MD", "city": "Baltimore", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "US", "state": "MI", "city": "Detroit", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "US", "state": "NJ", "city": "New Jersey", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "US", "state": "OH", "city": "Cleveland", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "US", "state": "OR", "city": "Portland", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "US", "state": "TX", "city": "Dallas", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "US", "state": "UT", "city": "Salt Lake City", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "US", "state": "VA", "city": "Ashburn", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "US", "state": "WA", "city": "Seattle", "provider_region": null, "self_serve": true, "choosable": true }, { "country": "VN", "state": null, "city": "Ho Chi Minh City", "provider_region": null, "self_serve": true, "choosable": true } ], "locations_source": "https://console.vast.ai/api/v0/bundles/ - marketplace offers with geolocation field", "locations_quote": "Vast.ai is a marketplace where hosts set their own rates and compete. Offers available across global distributed hosts with geolocation data via API.", "locations_checked": "2026-10-01" }