Skip to content

Mistral AI · Chat

Devstral Small 2505

Mistral AI 🇫🇷ChatDevstral familyOne of its sources says the weights aren't publicly available.
From NanoGPT's listing:
OpenHands+Devstral is 100% local 100% open, and is SOTA for the category on SWE-Bench Verified: 46.8% accuracy.
llminfo IQ
112.7
Maker's price
$0.075 blended / 1M
Max context
256K
Listings
15

Evidence and sources

🤗 chutesai/Devstral-Small-2505llminfo IQ 112.7, ranked among chat models. From 14 benchmarks. A fair amount of data behind it.How llminfo IQ is computedCoding IQ 82.9, 95% interval 73–93. From 3 benchmark families across 3 sources. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ. Its newest result is over a year old.How Coding IQ is computedPoor value (46.0). Costs more per task than its benchmark results predict. From 1 benchmark with measured task costs. Little data behind it, so treat it as a rough estimate.How Value for money worksReleased 21 May 2025Maker HQ: FR (according to Mistral AI's own website)
All rankings: #546 on NanoGPT recentlySources: Together AI data, Models.dev data, NanoGPT, llminfo researchPrimary data from Together AI from Models.dev: weights availability from NanoGPT: description from llminfo research: release date

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Trade-offs

No open weights

The weights aren't public, so you can't run it yourself.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 04:34 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 8 maker's prices; 1 market median price.

Input & output

as stated by the model's providers
Text
Images
Text

Capabilities

Supported operations and provider-stated features

VisionReasoningFunction calling

Listings

 Mistral AIdevstral-smallSource: GenAI Prices dataNot published by the provider$0.06$0.12
 Mistral AImistral/devstral-small-latest · latestSource: LiteLLM data256K$0.10cache read $0.01$0.30
 302aidevstral-small-2505 · 2505Source: 302.ai data128K$0.11$0.33
 Flowdotmistralai/devstral-smallSource: FlowDot data131K$0.10$0.30
 NanoGPTmistralai/devstral-small-2505 · 2505Sources: Models.dev data, NanoGPT33K$0.06cache read $0.03$0.06
 OpenRoutermistralai/devstral-small-2505 · 2505Source: Helicone Costs dataNot published by the provider$0.05$0.22
 OpenRouterdevstral-smallSource: GenAI Prices dataNot published by the provider$0.06$0.12
 RequestyInference by Mistral AImistral/devstral-small-latest · latestSource: Requesty data131K$0.11$0.33
 Vercelmistral/devstral-smallSource: Helicone Costs dataNot published by the provider$0.07$0.28
 Fireworks AIfireworks_ai/accounts/fireworks/models/devstral-small-2505 · 2505Source: LiteLLM data131K$0.90$0.90
 Ionetmistralai/Devstral-Small-2505 · 2505Source: Models.dev data128K$0.05cache read $0.025$0.22cache write $0.10
 Vercel AI Gatewayvercel_ai_gateway/mistral/devstral-smallSource: LiteLLM data128K$0.07$0.28

3 listings without a price are hidden.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Coding

SWE-rebenchSource: SWE-rebench

Reasoning

Around the web

  • devstral-2-123b-instruct-2512 Model by Mistral AI | NVIDIA NIM
    NVIDIA Explore Models Blueprints GPUs Docs ⌘K Ctrl+K ? Login mistralai mistralai/devstral-2-123b-instruct-2512 Experience Model Card …NVIDIA Explore Models Blueprints GPUs Docs ⌘K Ctrl+K ? Login mistralai mistralai/devstral-2-123b-instruct-2512 Experience Model Card Deploy This API will be deprecated on 05/11/2026. It will no longe…

    build.nvidia.com

  • devstral-2-123b-instruct-2512 Model by Mistral AI | NVIDIA NIM
    title: "devstral-2-123b-instruct-2512" publisher: "mistralai" type: "endpoint" updated: "2025-12-09T15:07:02.767Z" description: "State-of-the-art open code model with deep …title: "devstral-2-123b-instruct-2512" publisher: "mistralai" type: "endpoint" updated: "2025-12-09T15:07:02.767Z" description: "State-of-the-art open code model with deep reasoning, 256k context, an…

    build.nvidia.com

  • README.md · mistralai/Devstral-2-123B-Instruct-2512 at main
    software engineering tasks. Devstral 2 excels at using tools to explore codebases, editing multiple files …software engineering tasks. Devstral 2 excels at using tools to explore codebases, editing multiple files and power software engineering agents. The model achieves remarkable performance on SWE-bench…

    huggingface.co

  • README.md · mistralai/Devstral-Small-2505 at main - Hugging Face
    Key Features: Agentic coding : Devstral is designed to excel at agentic coding tasks, making …Key Features: Agentic coding : Devstral is designed to excel at agentic coding tasks, making it a great choice for software engineering agents. lightweight : with its compact size of just 24 billion …

    huggingface.co

  • README.md · mistralai/Devstral-2-123B-Instruct-2512 at main
    Devstral is an agentic LLM for software engineering tasks. Devstral 2 excels at using tools …Devstral is an agentic LLM for software engineering tasks. Devstral 2 excels at using tools to explore codebases, editing multiple files and power software engineering agents. The model achieves rema…

    huggingface.co

  • mistralai/Devstral-Small-2505 · Hugging Face
    mistralai/Devstral-Small-2505 · Hugging Face If it doesn't do it automatically, ask Devstral to deploy the …mistralai/Devstral-Small-2505 · Hugging Face If it doesn't do it automatically, ask Devstral to deploy the app or do it manually, and then go the front URL deployment to see the app. Now that you hav…

    huggingface.co

  • mistralai/Devstral-Small-2507 · Hugging Face
    llama.cpp Expand Download the weights from huggingface: pip install -U "huggingface_hub[cli]" huggingface-cli download "mistralai/Devstral-Small-2507_gguf" --include …llama.cpp Expand Download the weights from huggingface: pip install -U "huggingface_hub[cli]" huggingface-cli download "mistralai/Devstral-Small-2507_gguf" --include "Devstral-Small-2507-Q4_K_M.gguf"…

    huggingface.co

  • README.md · mistralai/Devstral-Small-2505_gguf at main
    Download the weights from huggingface: pip install -U "huggingface_hub[cli]" huggingface-cli download "mistralai/Devstral-Small-2505_gguf" --include "devstralQ4_K_M.gguf" --local-dir …Download the weights from huggingface: pip install -U "huggingface_hub[cli]" huggingface-cli download "mistralai/Devstral-Small-2505_gguf" --include "devstralQ4_K_M.gguf" --local-dir "mistralai/Devst…

    huggingface.co

  • mistralai (Mistral AI_)
    Pan's profile picture mistralai 's models 11 Sort: Recently updated mistralai/Devstral-Small-2507_gguf Text Generation · 24B …Pan's profile picture mistralai 's models 11 Sort: Recently updated mistralai/Devstral-Small-2507_gguf Text Generation · 24B · Updated Jul 23, 2025 · 4.3k · 49 System theme Company TOS Privacy About …

    huggingface.co

  • README.md · mistralai/Devstral-Small-2507 at 6c895c3bb1495411ee434c0ef7a1e1237119c1a0
    server to deploy Devstral Small 1.1 llama.cpp Expand OpenHands (recommended) Launch a server to deploy …server to deploy Devstral Small 1.1 llama.cpp Expand OpenHands (recommended) Launch a server to deploy Devstral Small 1.1 Make sure you launched an OpenAI-compatible server such as vLLM or Ollama as …

    huggingface.co

Also known as: devstral-small-2505, fireworks_ai/accounts/fireworks/models/devstral-small-2505, mistral/devstral-small, mistral/devstral-small-latest, mistralai/devstral-small, mistralai/devstral-small-2505, mistralai/devstral-small-2505:free, vercel_ai_gateway/mistral/devstral-small

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works