Mistral AI · AI model
Mistral Medium 3.5
128B dense multimodal model combining instruction following, reasoning and coding. Text and image input, 256k-token context, configurable reasoning effort. Open weights under a Modified MIT license; announced as the default model in Le Chat (now Mistral Vibe).
Last verified Available 7 sources
Key specs
- Provider
- Mistral AI
- Family
- Mistral Medium
- API identifier
- mistral-medium-3-5
- Availability
- Available
- Release date
- 2026-04-28
- License
- Open weights (Modified MIT)
- Context window
- 256,000 tokens
- API access
- Yes
- Consumer app
- Yes
Capabilities
Inputs
- Text input Yes
- Image input Yes
- Audio input No Model page lists only text and image as input modalities (no technical documentation published)
- Video input No Model page lists only text and image as input modalities (no technical documentation published)
- File input Yes PDF documents via Document QnA (OCR + model) in Chat Completions/Conversations; Document QnA is listed on the model page; uploaded files max 50 MB and 1,000 pages
Outputs
- Text output Yes
- Image output No Model page lists only text (with reasoning) output; image generation exists separately as a built-in agent tool
- Audio output No Model page lists only text (with reasoning) output
Features
- Reasoning Optional Adjustable via reasoning_effort ('none' or 'high'); 'high' recommended for agentic and code use cases
- Tool calling Yes Function Calling listed for /v1/chat/completions and /v1/conversations
- Web search Yes Built-in web_search / web_search_premium tools via the Conversations and Agents APIs (not Chat Completions); model page lists Built-In Tools
"Unknown" means we have not verified this capability yet; it does not mean the model lacks it.
Pricing
Estimate your monthly cost →| Dimension | Tier | Price | Since | Verified | Source |
|---|---|---|---|---|---|
| Input tokens | base | $1.50 per 1M tokens | 2026-10-12 | 2026-10-25 | API pricing | Mistral Docs Standard mode list price (Priority mode +75% and regional inference +10% not included) |
| Cached input tokens | base | $0.15 per 1M tokens | 2026-10-12 | 2026-10-25 | API pricing | Mistral Docs Cached input price listed on the docs pricing page (10% of the input price) |
| Output tokens | base | $7.50 per 1M tokens | 2026-10-12 | 2026-10-25 | API pricing | Mistral Docs Standard mode list price |
| Batch input tokens | base | $0.75 per 1M tokens | 2026-10-12 | 2026-10-25 | API pricing | Mistral Docs Batch API: 50% of standard price (docs pricing page Batch mode, factor 0.5: 'Batching ... at a 50% discount'; also stated on mistral.ai/pricing); computed from the standard list price |
| Batch output tokens | base | $3.75 per 1M tokens | 2026-10-12 | 2026-10-25 | API pricing | Mistral Docs Batch API: 50% of standard price (docs pricing page Batch mode, factor 0.5); computed from the standard list price |
Prices are shown in the currency the provider publishes; we never convert currencies.
Benchmarks
| Benchmark | Setting | Score | Reported by | Measured | Verified | Source |
|---|---|---|---|---|---|---|
| GDPval-AA v2.1 Elo, higher is better | v2.1; effort not stated; Stirrup harness (shell + web browsing) | 761 Elo | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Text Arena Arena score, higher is better | Overall, style control (default view); no effort suffix | 1,426 | Third party | 2026-10-23 | 2026-10-25 | Text Arena leaderboard (Arena) |
Elo is a relative rating: values change as models are added and the scale is re-anchored, so always store the date a rating was read. For example, Anthropic's launch page gives 1846 for Claude Opus 5.5, while the leaderboard on 2026-10-12 shows 1866 (Max, Default Fallback). Earlier versions (Google cites "GDPVal-AA v2") are not comparable. Judging is done by an LLM, not by human experts. Anthropic models are listed with "Default Fallback", meaning refused steps are routed to a fallback model. AA-Briefcase and GDP.pdf are separate evaluations.
Scores measure voter preference, which can be influenced by answer style and length. The leaderboard offers a style-control adjustment, so record which view was used. Scores shift with every update, so a value is only meaningful together with its leaderboard date. Entries marked "Preliminary" or with few votes have wide confidence intervals. The same model can appear several times with different reasoning-effort suffixes (for example -high or -max). Category views (Coding, Math, Creative Writing and others) are different rankings from Overall.
One row per result, without rank numbers: each result was measured under its own setting (effort level, harness, tools, test version), as stated by the source, and scores compare only within the same setting. Self-reported results come from the provider and have not been reproduced independently.
Rankings
- #30 of 42 in Best AI API as of 2026-10-26
- #32 of 39 in Best AI for business as of 2026-10-26
Derived from our published ranking snapshots; each ranking explains its criteria and weights.
Compare Mistral Medium 3.5
Guides
- AI token pricing explained: input, output, cached and batch rates
- Reasoning models explained: thinking modes, effort and cost
- Multimodal AI explained: image, audio, video and file inputs, and what they cost
- Context windows explained: limits, long-context pricing and recall
- How to choose an AI model: a practical decision framework
About Mistral Medium 3.5
Mistral Medium 3.5 is a 128B-parameter dense model that merges instruction following, reasoning and coding in a single set of weights. It accepts text and image input, has a 256k-token context window and supports configurable reasoning effort per request. Weights are published on Hugging Face under a Modified MIT license. In its launch announcement Mistral made it the default model in Le Chat (now Mistral Vibe) and the replacement for Devstral 2 in the Vibe CLI coding agent.
Strengths and weaknesses
Strengths
- Text and image input with configurable reasoning effort (
reasoning_effort) - Open weights (128B dense) under a Modified MIT license
- Supports structured outputs, function calling, predicted outputs, Document QnA, batch processing and built-in tools
Weaknesses
- Highest per-token list price in the flagship table of Mistral's API pricing page ($1.50 input / $7.50 output per 1M tokens)
- 256k-token context window, smaller than Mistral Large 4 (1M)
- No audio or video input
Sources
Last verified Page updated 2026-10-26