Find your next compute setup.
288 models · 41 GPUs
- DeepSeek V4 FlashAPI model →
- GPT-OSS 120BAPI model →
- GLM-5.2API model →
- GLM-5.3API model →
- DeepSeek V4.1 FlashAPI model →
- GLM-5.3 FlashAPI model →
- DeepSeek V4 ProAPI model →
- Kimi K2.7 CodeAPI model →
- Kimi K2.6API model →
- Kimi K3API model →
- Llama 3.3 70BAPI model →
- DeepSeek R1API model →
- GPT-OSS 20BAPI model →
- DeepSeek V3API model →
- MiniMax M3API model →
- GLM-5.1API model →
- Llama 3.1 8BAPI model →
- Claude Haiku 4.5API model →
- Claude Sonnet 5API model →
- DeepSeek-V3.1API model →
- Gemma 4 31BAPI model →
- Qwen3.6 35B A3BAPI model →
- Kimi K3.1 ProAPI model →
- Qwen3.8 Max (0902)API model →
- Qwen3.8 27BAPI model →
- Claude Fable 5API model →
- Claude Opus 5API model →
- Claude Sonnet 4.6API model →
- Kimi K2.5API model →
- Qwen3 Next 80B A3B InstructAPI model →
- Qwen3 235B A22BAPI model →
- GLM-5API model →
- MiniMax M3.1 FlashAPI model →
- Qwen3.5 397B A17BAPI model →
- Claude Opus 4.7API model →
- Claude Opus 5.5API model →
- Claude Sonnet 4.5API model →
- GPT-5.6 LunaAPI model →
- Llama 4 ScoutAPI model →
- DeepSeek V3.2API model →
- MiniMax M2.5API model →
- Kimi K2 ThinkingAPI model →
- Llama 3.1 70B InstructAPI model →
- Qwen3 30B A3BAPI model →
- Gemma 4 26B A4B API model →
- Qwen3-Next-80B-A3B-ThinkingAPI model →
- Qwen3.8 FlashAPI model →
- Kimi K2API model →
- GLM-4.7API model →
- Claude Haiku 5.5API model →
- MiniMax M2.7API model →
- Qwen3.6 27BAPI model →
- Claude Opus 4.5API model →
- Claude Opus 4.6API model →
- Claude Opus 4.8API model →
- GPT-5.6 TerraAPI model →
- Llama 4 MaverickAPI model →
- Grok 4.6API model →
- Mistral NemoAPI model →
- Claude Sonnet 5.5API model →
- Llama 3.2 3BAPI model →
- Qwen3 VL 235B A22B InstructAPI model →
- Qwen3.7 MaxAPI model →
- Gemini 3.1 Pro PreviewAPI model →
- Gemini 3.7 FlashAPI model →
- DeepSeek R1 Distill Llama 70BAPI model →
- GLM-4.5 AirAPI model →
- GLM-4.6API model →
- Claude Fable 5.1API model →
- GPT-5.6 SolAPI model →
- GPT-5.4API model →
- GPT-5API model →
- GPT-5 MiniAPI model →
- GLM-4.7 FlashAPI model →
- Gemini 2.5 FlashAPI model →
- Gemini 3 Flash PreviewAPI model →
- Gemini 3.5 FlashAPI model →
- Qwen3 32BAPI model →
- Claude Fable 5.2API model →
- Grok 4.7API model →
- MiMo-V2.6-FlashAPI model →
- Hy4API model →
- Gemini 4 ArgonAPI model →
- Claude-Sonnet-4API model →
- GPT-6 AstraAPI model →
- GPT-6 LunaAPI model →
- GPT-5.4 NanoAPI model →
- GPT-5 NanoAPI model →
- GPT-5.2API model →
- Mistral LargeAPI model →
- Mistral Small 4API model →
- Gemini 2.5 ProAPI model →
- Gemini 3.8 FlashAPI model →
- Gemini 3.5 Flash LiteAPI model →
- Qwen3-Coder-480b-A35b-InstructAPI model →
- GLM-4.5API model →
- Qwen3.7 PlusAPI model →
- Nemotron 3 UltraAPI model →
- Qwen3.5-122B-A10BAPI model →
- MiMo-V2.6-ProAPI model →
- Hy3API model →
- Step 3.7 FlashAPI model →
- Gemini 4 CarbonAPI model →
- Claude-3-5-SonnetAPI model →
- Claude-3.7-SonnetAPI model →
- Claude-4-OpusAPI model →
- GPT-6.1 SolAPI model →
- GPT-5.5API model →
- GPT-5.4 MiniAPI model →
- GPT-4oAPI model →
- GPT-4o-MiniAPI model →
- GPT-5.1API model →
- GPT-4.1API model →
- GPT-4.1 MiniAPI model →
- Grok 4.3API model →
- DeepSeek R1 Distill Qwen 32BAPI model →
- QwQ-32BAPI model →
- Gemini 3.1 Flash LiteAPI model →
- Gemini 3.6 FlashAPI model →
- InklingAPI model →
- Gemma 3 27BAPI model →
- Nemotron 3.5 Super VLAPI model →
- MiniMax M2API model →
- Qwen2.5 VL 72B InstructAPI model →
- Qwen3 Coder 30B A3B InstructAPI model →
- Muse Glimmer 30BAPI model →
- DeepSeek V4 Flash Vision ExpAPI model →
- Muse Spark OpenAPI model →
- MiMo-V2.5API model →
- Qwen3.5 27BAPI model →
- Qwen3.5 35B A3BAPI model →
- Claude-3.5-HaikuAPI model →
- Llama-3-70bAPI model →
- Llama-3-8bAPI model →
- Mixtral-8x7B-Instruct-V0.1API model →
- GPT-6 SolAPI model →
- GPT-6.1 TerraAPI model →
- GPT-4.1 NanoAPI model →
- O3API model →
- Codestral 2508API model →
- Grok-4-20-ReasoningAPI model →
- Grok-4-20-Non-ReasoningAPI model →
- MiniMax M2.1API model →
- Llama-3.2-1bAPI model →
- Qwen2p5-Coder-32bAPI model →
- Qwen3-VL-235b-A22b-ThinkingAPI model →
- Qwen3-VL-32b-InstructAPI model →
- Llama 4 MaverickAPI model →
- Qwen2p5-72bAPI model →
- Qwen3 14BAPI model →
- Qwen3 235B A22B Thinking 2507API model →
- DeepSeek V3.1 TerminusAPI model →
- Gemma 3 12BAPI model →
- Gemini 2.5 Flash LiteAPI model →
- Llama 3.1 405BAPI model →
- Muse Spark 1.2API model →
- Qwen3.5-9BAPI model →
- Qwen3.8 2.4T A95BAPI model →
- Grok 4.5API model →
- Qwen3 VL 8B InstructAPI model →
- MiMo-V2.5-ProAPI model →
- Step 5 PreviewAPI model →
- Gemini 4 HydrogenAPI model →
- Claude Opus 4.1API model →
- O1API model →
- O3-MiniAPI model →
- GPT-3.5 Turbo (older v0613)API model →
- O4-MiniAPI model →
- Grok-4API model →
- Ministral 3 3B 2512API model →
- Qwen4 MaxAPI model →
- Qwen3.8 Omni FlashAPI model →
- Nemotron 3.5 LightningAPI model →
- Pixtral Large 2502API model →
- DeepSeek Flash LatestAPI model →
- Muse Spark 1.1API model →
- Muse Spark 1.2 ContributorAPI model →
- Muse Spark 1.3API model →
- Muse Spark 1.3 ContributorAPI model →
- GLM LatestAPI model →
- Mistral Large 4API model →
- GLM 5.3 FlashXAPI model →
- Ling 3.0 FlashAPI model →
- Granite 4.2 8BAPI model →
- Inkling SmallAPI model →
- Gemini Omni 1.1 FlashAPI model →
- GLM Flash LatestAPI model →
- Motif 3API model →
- Ember-1API model →
- ExtractAPI model →
- SummarizeAPI model →
- Beam 1.0 501B A23BAPI model →
- GPT-5.6 Luna ProAPI model →
- GPT-5.6 Sol ProAPI model →
- GPT-5.6 Terra ProAPI model →
- Muse WonderAPI model →
- Qwen3.7 FlashAPI model →
- Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)API model →
- Fugu MaxAPI model →
- MiMo-V2.6-Pro-UltraSpeedAPI model →
- Seed 2.1 TurboAPI model →
- Ripple 65B PreviewAPI model →
- Gemini Omni 1.2 FlashAPI model →
- Inkling 2 SmallAPI model →
- GPT-6.2 AstraAPI model →
- GPT-6.2 SolAPI model →
- Gemini 4 NeonAPI model →
- Inkling 2 MiniAPI model →
- DaisyRose 27BAPI model →
- Gemini 4 Neon (batch)API model →
- Hy4 flashAPI model →
- Gemini 4 Hydrogen (batch)API model →
- ERNIE-Light-DT-1.0API model →
- Gemini 4 Carbon (batch)API model →
- Nex-N2.6API model →
- Claude Fable 5.2 (batch)API model →
- Muse EmberAPI model →
- LFM3-1.7BAPI model →
- Mi:dm K 3.0 ProAPI model →
- GLM 5.3 FlashX LegalAPI model →
- Particle-1API model →
- Kimi K3.1 Pro (batch)API model →
- DeepSeek V4.1 Flash 1017 (batch)API model →
- GlimpseAPI model →
- Ternary Bonsai 2 9BAPI model →
- Moondream 3 PreviewAPI model →
- Moondream 3.1API model →
- Pareto 26.10API model →
- Pareto LatestAPI model →
- Pareto 26.9API model →
- LFM2.5-24B-A2BAPI model →
- Gemini 4 Argon (batch)API model →
- Nano Banana 2.1API model →
- Ling 3.1 FlashAPI model →
- Claude Sonnet 5.5 (batch)API model →
- Perceptron Mk1.5API model →
- GLM 5.3 PrimeAPI model →
- Qwen3.8 Max PrimeAPI model →
- Aion 3.5 MiniAPI model →
- Aion 3.5API model →
- Solar Mini 4API model →
- Command A+API model →
- Claude Opus 5.5 (batch)API model →
- Ternary Bonsai 2 27BAPI model →
- DeepSeek Pro LatestAPI model →
- Schematron V2 TurboAPI model →
- Schematron V2 SmallAPI model →
- Ling 3.0 Flash VLAPI model →
- DeepSeek V4.1 Flash (batch)API model →
- Nex-N2.5-MiniAPI model →
- Nex-N2.5-ProAPI model →
- Qwen3.8 Max (0902) (batch)API model →
- Gemini 3.8 Flash (batch)API model →
- Claude Fable 5.1 (batch)API model →
- GLM 5.3 Flash (batch)API model →
- Hy-MT2-1.8BAPI model →
- Hy-MT2-30B-A3BAPI model →
- Hy-MT2-7BAPI model →
- GLM 5.3 (batch)API model →
- Qwen3.8 27B (batch)API model →
- Gemini 3.7 Flash (batch)API model →
- Sakana NamazuAPI model →
- Solar Pro 4API model →
- Muse Glimmer 30B (batch)API model →
- Ling 3.0 TinyAPI model →
- DeepSeek V4 Flash 0731 (batch)API model →
- Inkling Small (batch)API model →
- Claude Opus 5 (batch)API model →
- Gemini 3.6 Flash (batch)API model →
- Gemini 3.5 Flash Lite (batch)API model →
- Kimi K3 (batch)API model →
- Aion-3.0-MiniAPI model →
- Aion-3.0API model →
- Claude Sonnet 5 (batch)API model →
- GLM 5.2 (batch)API model →
- Claude Fable LatestAPI model →
- Claude Fable 5 (batch)API model →
- Nemotron 3.5 Content SafetyAPI model →
- Qwen3.7 Plus (batch)API model →
- MiniMax M3 (batch)API model →
- Claude Opus 4.8 (batch)API model →
- Qwen3 CoderAPI model →
- Mistral Medium 3.5API model →
- Nemotron 3 SuperAPI model →
- Nemotron 3 NanoAPI model →
- IBM Granite 4.1 8BAPI model →
- Phi-4API model →
- Mixtral 8x7BAPI model →
- A6000GPU →
- RTX PRO 6000GPU →
- 6000 AdaGPU →
- L40SGPU →
- A100 SXM 80GBGPU →
- H100 SXMGPU →
- H200GPU →
- B200GPU →
- B300GPU →
- V100GPU →
- A100 SXM 40GBGPU →
- RTX A2000GPU →
- RTX A5000GPU →
- RTX A4000GPU →
- RTX 4070 TiGPU →
- RTX A4500GPU →
- RTX 4000 Ada GenerationGPU →
- RTX 3090GPU →
- RTX 2000 Ada GenerationGPU →
- RTX 4080GPU →
- RTX 4090GPU →
- RTX PRO 4500 BlackwellGPU →
- A40GPU →
- RTX 5080GPU →
- L4GPU →
- RTX 5000 Ada GenerationGPU →
- MI300XGPU →
- MI350 OAMGPU →
- H200 NVLGPU →
- RTX PRO 4000 BlackwellGPU →
- RTX PRO 4500 Blackwell Server EditionGPU →
- RTX 5090GPU →
- L40GPU →
- RTX PRO 5000 BlackwellGPU →
- A100 PCIeGPU →
- H100 PCIeGPU →
- H100 NVLGPU →
- GB300GPU →
- RTX 3070GPU →
- RTX 3080GPU →
- RTX 3080 TiGPU →
Three ways to run your workload
Try a one-click cost comparison → · API vs rental calculator · Self-hosting estimates
Before you rent, check the whole bill.
Considering a GPU rental? See a sample workload decision brief and tell us whether a $49 one-time review would help. No payment or order. Explore the pilot →
Tools, conversations & models
GitHub tools
Snapshot 2026-11-22 UTC
More tools (5)
Discussions · Hacker News
Fetched 2026-11-22 UTC
- Frostbite: A new CPU architecture discovered, designed and verified by AI
- Restricting AI agents in mobile operating systems platforms from capturing data
- Nasdaq erases all 2026 gains as AI sell-off dampens Anthropic’s second day
More discussions (5)
- AI agents ask better questions than KPMG accountants in audit trial
- Language Models Can Rewrite Their Own Weights
- DreamCoder: Coding harness and IDE from the author of Mastering Emacs
- Pete Warden on local AI, open models, voice interfaces, the future of hardware
- LLMs can read and write executables byte by byte from a hex dump
Saved X posts
Snapshot 2026-11-14 UTC · archive
- 🚀 The full DeepSeek App experience, now via API Search, files, links, safety, and voice. The capabilities behind the DeepSeek App are now available through the DeepSeek API. Enable built-in features with a few paramete
- llama.cpp v1.0.0 is now available
- We’re not going to win by lobbying, plotting, or manipulating. We’re going to win by building, in the open, with the whole community! Today, with @ggerganov we’re super excited to announce the release of llama.cpp v1, t
Hugging Face models
Snapshot 2026-11-22 UTC
A snapshot of the market
Listed prices, not performance rankings. Check availability and terms with the provider before buying.
API prices · per 1M tokens
all 288 →Both prices are from one provider selected by our price-check rules. Output may be cheaper elsewhere.
| Model | Input | Output | Provider |
|---|---|---|---|
| Pixtral Large 25024 providers · listed 2026-11-20 | $2.00 | $6.00 | AWS Bedrock |
| Gemini 4 Neon4 providers · listed 2026-11-18 | $3.00 | $15.00 | Google Gemini API |
| GPT-6.2 Astra3 providers · listed 2026-11-17 | $8.00 | $40.00 | OpenAI |
| GPT-6.2 Sol3 providers · listed 2026-11-17 | $2.00 | $10.00 | OpenAI |
| Qwen4 Max3 providers · listed 2026-11-16 | $1.05 | $3.15 | Together AI |
GPU rental · per hour
all 41 →Prices are per GPU-hour. Spot and community offers have different availability and terms.
New this week
all new models →- Inkling 2 Mini listed 2026-11-21 · in $0.075
- DaisyRose 27B listed 2026-11-21 · in $0.150
- Pixtral Large 2502 listed 2026-11-20 · in $2.00
- Gemini 4 Neon listed 2026-11-18 · in $3.00
- Gemini 4 Neon (batch) listed 2026-11-18 · in $1.50
Some provider links are referral links. They do not affect price ordering. How we earn revenue.