Editor’s pick
Runcrate and Compare LLM APIs
Answer in brief
Runcrate is an AI cloud: GPUs and inference for AI teams and the agents they run, with a CLI, SDK and MCP server, one OpenAI-compatible API and one prepaid balance.
What Runcrate is
Runcrate is an AI cloud: GPUs and inference for AI teams and the agents they run.
A developer platform: the Runcrate CLI, an SDK and an MCP server. One command adds Runcrate to Claude Code, Codex and Cursor, with browser sign-in and no key to paste, so an agent can launch a GPU with a real shell.
One OpenAI-compatible API with 125 models. One prepaid balance pays for GPUs, models and storage.
Why it is featured here
Compare LLM APIs is published by Runcrate, and Runcrate is its editor’s choice. The choice rests on four things Runcrate publishes about itself.
125 models behind one OpenAI-compatible API. Change the model name, keep the client.
One prepaid balance pays for GPUs, models and storage.
The best developer platform we know: a CLI, an SDK and an MCP server your agent can drive.
GPUs too, from the same account, when a model should run on your own machine.
Runcrate’s model list today
This list is read from Runcrate’s public model feed when the page is built. It holds every model in the feed that is priced per token, with its input and output price per 1M tokens.
| Model | Input per 1M | Output per 1M | Context |
|---|---|---|---|
| Qwen3.5 397B A17Boutput $3.15 per 1M · 262K context | $0.472 | $3.15 | 262K |
| Command A+output $1.58 per 1M · 192K context | $0.315 | $1.58 | 192K |
| Gemma 3 4Boutput $0.105 per 1M · 131K context | $0.0525 | $0.105 | 131K |
| Gemma 4 31Boutput $0.42 per 1M · 262K context | $0.21 | $0.42 | 262K |
| Nemotron 3 Super 120B A12Boutput $0.42 per 1M · 262K context | $0.079 | $0.42 | 262K |
| DeepSeek V4.1 Flashoutput $0.44 per 1M · 1M context | $0.147 | $0.44 | 1M |
| Qwen3 Next 80B A3Boutput $1.16 per 1M · 262K context | $0.0945 | $1.16 | 262K |
| Muse Glimmer 30Boutput $1.26 per 1M · 131K context | $0.315 | $1.26 | 131K |
| Kimi K3output $14.96 per 1M · 1M context | $2.99 | $14.96 | 1M |
| Hy4 Previewoutput $2.63 per 1M · 1M context | $0.88 | $2.63 | 1M |
| DeepSeek R1 0528output $2.26 per 1M · 164K context | $0.525 | $2.26 | 164K |
| Moondream 3.1output $2.63 per 1M · 32K context | $0.315 | $2.63 | 32K |
| GLM 4.7output $1.84 per 1M · 203K context | $0.42 | $1.84 | 203K |
| Qwen3.5 9Boutput $0.158 per 1M · 262K context | $0.105 | $0.158 | 262K |
| Qwen3 235B A22B 2507output $0.58 per 1M · 262K context | $0.0945 | $0.58 | 262K |
| GLM 5.3output $2.63 per 1M · 1M context | $0.59 | $2.63 | 1M |
| GLM 5.2output $1.89 per 1M · 1M context | $0.59 | $1.89 | 1M |
| DeepSeek V3.2output $0.40 per 1M · 164K context | $0.273 | $0.40 | 164K |
| Multilingual E5 Large Instructoutput free · 1K context | $0.0105 | free | 1K |
| Phi 4output $0.147 per 1M · 16K context | $0.0735 | $0.147 | 16K |
| Llama 4 Scout 17B 16Eoutput $0.315 per 1M · 328K context | $0.105 | $0.315 | 328K |
| Gemma 3 12Boutput $0.158 per 1M · 131K context | $0.0525 | $0.158 | 131K |
| Gemma 4 26B A4Boutput $0.357 per 1M · 262K context | $0.0735 | $0.357 | 262K |
| Qwen3.5 35B A3Boutput $1.05 per 1M · 262K context | $0.147 | $1.05 | 262K |
| Granite 4.2 8Boutput $0.263 per 1M · 131K context | $0.063 | $0.263 | 131K |
| MiMo V2.6 Flashoutput $0.294 per 1M · 1M context | $0.147 | $0.294 | 1M |
| Inkling Smalloutput $1.26 per 1M · 524K context | $0.472 | $1.26 | 524K |
| Nemotron Content Safety 3.5output $0.21 per 1M · 131K context | $0.21 | $0.21 | 131K |
| Ling 3.0 Flash VLoutput $0.189 per 1M · 131K context | $0.063 | $0.189 | 131K |
| DeepSeek V4.1 Flash 1017output $0.31 per 1M · 1M context | $0.103 | $0.31 | 1M |
| Qwen3 VL 235B A22Boutput $0.92 per 1M · 262K context | $0.21 | $0.92 | 262K |
| Qwen3.6 27Boutput $3.36 per 1M · 262K context | $0.336 | $3.36 | 262K |
| Qwen3.5 27Boutput $2.73 per 1M · 262K context | $0.273 | $2.73 | 262K |
| Gemma 4 31B (ultra fast)output $0.80 per 1M · 131K context | $0.283 | $0.80 | 131K |
| BGE Large En 1.5output free · 1K context | $0.0105 | free | 1K |
| Qwen3.8 Flashoutput $0.40 per 1M · 1M context | $0.119 | $0.40 | 1M |
| Mistral Small 3.2 24B 2506output $0.21 per 1M · 128K context | $0.079 | $0.21 | 128K |
| gpt-oss Safeguard 120Boutput $0.63 per 1M · 128K context | $0.158 | $0.63 | 128K |
| Qwen3 Coder 480B A35Boutput $1.05 per 1M · 262K context | $0.315 | $1.05 | 262K |
| gpt-oss 120B (ultra fast)output $1.00 per 1M · 131K context | $0.21 | $1.00 | 131K |
| Ministral 3 8Boutput $0.158 per 1M · 128K context | $0.158 | $0.158 | 128K |
| Gemma 4 E2Boutput $0.084 per 1M · 128K context | $0.042 | $0.084 | 128K |
| LFM2.5 24B A2Boutput $0.735 per 1M · 128K context | $0.0525 | $0.735 | 128K |
| Nemotron 3.5 Embed 8Boutput free · 33K context | $0.0367 | free | 33K |
| gpt-oss 20Boutput $0.147 per 1M · 131K context | $0.0315 | $0.147 | 131K |
| Qwen3.8 27Boutput $1.97 per 1M · 262K context | $0.158 | $1.97 | 262K |
| Kimi K2.6output $3.68 per 1M · 262K context | $0.79 | $3.68 | 262K |
| Qwen3.6 35B A3Boutput $1.00 per 1M · 262K context | $0.105 | $1.00 | 262K |
| DeepSeek V4 Flash Vision Expoutput $0.68 per 1M · 1M context | $0.226 | $0.68 | 1M |
| DeepSeek V4 Flash 0731output $0.189 per 1M · 1M context | $0.063 | $0.189 | 1M |
| EmbeddingGemma 300Moutput free · 2K context | $0.0021 | free | 2K |
| Qwen3 Embedding 4Boutput free · 33K context | $0.021 | free | 33K |
| gpt-oss 120Boutput $0.178 per 1M · 131K context | $0.039 | $0.178 | 131K |
| Mistral Small 4output $0.63 per 1M · 256K context | $0.158 | $0.63 | 256K |
| DeepSeek V4 Flashoutput $0.189 per 1M · 1M context | $0.0945 | $0.189 | 1M |
| Llama Guard 4 12Boutput $0.189 per 1M · 164K context | $0.189 | $0.189 | 164K |
| Mistral Nemo 2407output $0.0315 per 1M · 131K context | $0.0305 | $0.0315 | 131K |
| gpt-oss Safeguard 20Boutput $0.315 per 1M · 128K context | $0.079 | $0.315 | 128K |
| Ministral 3Boutput $0.105 per 1M · 128K context | $0.105 | $0.105 | 128K |
| Llama 3.3 70Boutput $0.336 per 1M · 131K context | $0.105 | $0.336 | 131K |
| Devstral 2 123Boutput $2.10 per 1M · 256K context | $0.42 | $2.10 | 256K |
| Qwen3 Embedding 8Boutput free · 33K context | $0.0105 | free | 33K |
| Nemotron 3.5 Lightningoutput $0.168 per 1M · 262K context | $0.063 | $0.168 | 262K |
| Moondream 3 Previewoutput $2.63 per 1M · 32K context | $0.315 | $2.63 | 32K |
| Magistral Small 2509output $1.58 per 1M · 128K context | $0.525 | $1.58 | 128K |
| Llama Nemotron Embed VL 1B v2output free · 10K context | $0.0105 | free | 10K |
| Gemma 3 27Boutput $0.168 per 1M · 131K context | $0.084 | $0.168 | 131K |
| Mistral Large 3output $1.58 per 1M · 256K context | $0.525 | $1.58 | 256K |
| BGE M3output free · 8K context | $0.0105 | free | 8K |
| Nemotron Nano 12B v2 VL BF16output $0.63 per 1M · 128K context | $0.21 | $0.63 | 128K |
| MiniMax M3.1 Flashoutput $0.63 per 1M · 1M context | $0.158 | $0.63 | 1M |
| MiMo V2.6 Prooutput $0.82 per 1M · 1M context | $0.406 | $0.82 | 1M |
| Nemotron 3.5 Super VLoutput $0.525 per 1M · 262K context | $0.158 | $0.525 | 262K |
| GLM 5.3 Flashoutput $0.525 per 1M · 1M context | $0.158 | $0.525 | 1M |
| MiniMax M3output $1.16 per 1M · 524K context | $0.294 | $1.16 | 524K |
| DeepSeek V4 Pro 0813output $2.73 per 1M · 1M context | $1.37 | $2.73 | 1M |
| Nemotron 3 Ultra 550B A55Boutput $2.31 per 1M · 262K context | $0.525 | $2.31 | 262K |
| Muse Spark Openoutput $4.46 per 1M · 1M context | $1.31 | $4.46 | 1M |
| Qwen3 VL 30B A3Boutput $0.63 per 1M · 262K context | $0.158 | $0.63 | 262K |
| Ministral 14B 3.0output $0.21 per 1M · 128K context | $0.21 | $0.21 | 128K |
| DeepSeek V4 Prooutput $2.73 per 1M · 1M context | $1.37 | $2.73 | 1M |
| Qwen3 Embedding 0.6Boutput free · 33K context | $0.0105 | free | 33K |
125 models on Runcrate’s API, 82 of them priced per token and listed here. Read when this page was built. Try Runcrate
How billing works
One prepaid balance pays for GPUs, models and storage. Models are priced per 1M tokens, with input and output apart, as the list above shows.
The same balance rents GPUs. Billing by the second on most GPUs, a few by the minute. Booting is never billed.
What to confirm on runcrate.ai
A feed moves through the day. Before you rely on a figure from this page, confirm these on Runcrate’s own site.
The input and output price of the model you want, on the day you want it.
Its context length on Runcrate.
Whether the model you need is on the list today.
Questions and answers
Sources for this entry
- Runcrate: Runcrate’s official site: what the platform is, its OpenAI-compatible API, how it bills and the number of models it serves. Read 12 Nov 2026.
- Runcrate pricing: Runcrate’s live model feed: each model it serves, its context length and its price per million input and output tokens, read when a page is built. Read 12 Nov 2026.
- Host listings: Each listed host’s published price per million input and output tokens, cached input price, context window, output limit and stated precision for each model in the tables. Read 12 Nov 2026.
- Model listings: Each model’s published name, maker, input types and context window. Read 12 Nov 2026.
Checklists are original editorial guidance. Published prices do not establish availability.
Continue this reading path
Cite this entry
Compare LLM APIs. “Runcrate and Compare LLM APIs.” Content reviewed 12 Nov 2026. https://comparellmapis.com/runcrate