Mistral Large 4 at a glance
Mistral Large 4 launched as a public API preview on October 6, 2026, with open weights following later that month under a modified MIT license.
Current standing
As of
- Rank
- #30of 43 models
Mistral Large 4 (High) · 26.1 on the Artificial Analysis Intelligence Index v5.0.1 (opens in a new tab)
- Recorded at publication
- 26 (v5.0.1)
as of
- Current leader
- Gemini 4 Carbon (Google) · 60
Computed from our daily snapshot; it changes as new models are measured. Mistral previews Large 4, a 1T-parameter multimodal model, then opens its weights
What changed with Mistral Large 4
Mistral AI opened a public preview of Mistral Large 4 on October 6, 2026: a natively multimodal model with about 1 trillion total and 52 billion active parameters, strong at document and chart analysis. Pricing is $0.70 input, $0.07 cached input, and $2.10 output per 1M tokens.
A later Mistral post (dated October 29, 2026 on Mistral's news index) released the weights on Hugging Face under the Mistral Ecosystem Open-Weight license, a modified MIT license that requires a separate commercial license for model-as-a-service providers. It is also offered via Mistral's API, OpenRouter, NVIDIA endpoints, and Microsoft Foundry.
Mistral Large 4 benchmarks and pricing
Values are copied from the linked source on the date shown. They are not IA Compa measurements.
- Artificial Analysis Intelligence Indexas of · Source for Artificial Analysis Intelligence Index: artificialanalysis.ai (opens in a new tab)
- 26 (v5.0.1)
- Output speedas of · Source for Output speed: artificialanalysis.ai (opens in a new tab)
- 113 tokens/s
- Input price per 1M tokensas of · Source for Input price per 1M tokens: mistral.ai (opens in a new tab)
- $0.70
- Output price per 1M tokensas of · Source for Output price per 1M tokens: mistral.ai (opens in a new tab)
- $2.10
- Context windowas of · Source for Context window: artificialanalysis.ai (opens in a new tab)
- 1M tokens
How it compares
Other models in the same category that we have covered, using each model's latest recorded value. (Models)
Mistral Large 4 reasoning effort
Artificial Analysis tests this model at several reasoning-effort settings. Higher effort usually scores higher but costs more and answers more slowly.
Mistral Large 4 by effort level
Higher effort lets the model think longer. Here is what each setting scores, how fast it writes, and what it costs on Artificial Analysis.
| Effort | Intelligence Index | Output speed (tokens per second) | Cost to run the index | Blended price (3:1) (per 1M tokens) | Price in / out (per 1M tokens) | First token |
|---|---|---|---|---|---|---|
| High (Mistral Large 4 (high)) · Best: highest effort with a score, used in the leaderboard | 26 | 113 | $1,360 | $1.05 | $0.70 / $2.10 | 3.1 s |
| Medium (Mistral Large 4 (medium)) | 23 | 103 | $936 | $1.05 | $0.70 / $2.10 | 2.8 s |
Benchmarked by AA: high · medium of 3 available
Where to compare Mistral Large 4
Independent leaderboards and official resources for comparing this model yourself. Rankings change often; each site shows its own current data.
Leaderboards
General leaderboards
Why Mistral Large 4 matters for businesses
Artificial Analysis lists it at 26 on Intelligence Index v5.0.1 with about 113 output tokens per second. For European clients or anyone who must self-host, it is one of the largest open-weight options, at a low hosted price.
Mistral Large 4 updates since publication
Dated changes since publication. The original article text above is kept as written on its date.
Update (Updated)
The open weights released on October 29, 2026 are an improved checkpoint with three more weeks of reinforcement learning. Mistral reports 41% on Terminal-Bench 4.0, up from 28% in the preview. Reasoning modes are Low (replacing None), Medium and High, with High as the default.
mistral.ai (opens in a new tab)
Sources for this story
- Introducing Mistral Large 4 (opens in a new tab)Official
Mistral AI · mistral.ai
- Le Chonk: Built for Coders, Trusted by Defenders (opens in a new tab)Official
Mistral AI · mistral.ai
- Mistral Large 4 model analysis (opens in a new tab)Benchmark
Artificial Analysis · artificialanalysis.ai
Last verified: . IA Compa summarizes public announcements for business readers. We are not affiliated with the companies mentioned. Always confirm details, pricing, and availability with the official source.
- #llm
- #mistral
- #open-weights
- #multimodal
