Mistral AI model product
Mistral Large 3
Mistral Large 3 is a state-of-the-art general-purpose Multimodal granular Mixture-of-Experts model with 41B active parameters and 675B total parameters trained from scratch with 3000 H200s.
Updated Aug 12, 2026. Default version: Mistral Large 3 (675B Instruct 2512)
LLMBoard score15.8Mistral Large 3 (675B Instruct 2512)
Coverage80%5 benchmark families
Context window262.1KTokens
Official input price$0.50Mistral API
Mistral Large 3 Specifications
Technical details for the model's default version.
- Version
- Mistral Large 3 (675B Instruct 2512)
- Released
- Dec 4, 2025
- Knowledge cutoff
- Nov 1, 2024
- Parameters
- 675B
- Context window
- 262.1K
- Max output
- 262.1K
- Inputs
- image, text
- Outputs
- text
- Open weights
- Yes
- License
- Apache 2.0
Mistral Large 3 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Mistral Large 3 Benchmark Results
Benchmark scores for Mistral Large 3 (675B Instruct 2512).
| AMC_2022_23 | 52.0% | 04 | 6 | 40.0% | C | |
| MMMLU | 85.5% | 31 | 49 | 37.5% | C | |
| SimpleQA | 23.8% | 33 | 46 | 28.9% | C | |
| LiveCodeBench | 34.4% | 57 | 73 | 22.2% | C | |
| GPQA | 43.9% | 196 | 234 | 16.3% | C |
Mistral Large 3 Arena Results
Preference and agent-evaluation results for the default version.
No Arena results
The default version does not have a matching Arena result yet.
Mistral Large 3 Runtime Performance
Provider-specific output speed and catalog latency for Mistral Large 3 (675B Instruct 2512). Runtime does not affect the capability score.
| Mistral AI | 169.143 tok/s | 1.308 s | 262.1K | 262.1K |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Mistral Large 3 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $0.50 input, $1.5 output per 1M
- Official provider
- Mistral
- Lowest third-party
- From $0.50 input, $1.5 output per 1M via Merge Gateway
- Tracked offerings
- 4
| Nvidia | mistralai/mistral-large-3-675b-instruct-2512 | global | N/A | N/A | 262.1K | |
| Mistral | mistral-large-latest | global | $0.50 | $1.5 | 262.1K | |
| Merge Gateway | mistral/mistral-large-latest | global | $0.50 | $1.5 | 262.1K | |
| LLM Gateway | mistral-large-latest | global | $4 | $12 | 128K |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Mistral Large 3 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
| Mistral Large 3 (675B Instruct 2512 Eagle) | N/A | 675B | N/A | N/A | No | Apache 2.0 | |
| Mistral Large 3 (675B Base) | N/A | 675B | N/A | N/A | No | Apache 2.0 | |
| Mistral Large 3 (675B Instruct 2512 NVFP4) | N/A | 675B | N/A | N/A | No | Apache 2.0 | |
| Mistral Large 3 (675B Instruct 2512) | 15.8 | 675B | 262.1K | 262.1K | Yes | Apache 2.0 | |
| Mistral Large 3 | N/A | 675B | 262.1K | 262.1K | Yes | Apache 2.0 |
What is Mistral Large 3?
Key information about Mistral Large 3 and its available data.
Mistral Large 3 is a state-of-the-art general-purpose Multimodal granular Mixture-of-Experts model with 41B active parameters and 675B total parameters trained from scratch with 3000 H200s.
This model is the instruct post-trained version, fine-tuned for instruction tasks, making it ideal for chat, agentic and instruction based use cases.
Designed for reliability and long-context comprehension - It is engineered for production-grade assistants, retrieval-augmented systems, scientific workloads, and complex enterprise workflows.
Data as of 2026-08-11.
Mistral Large 3 vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to Mistral Large 3
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
MA
Mistral Small 3.2 24B
Mistral AI
MA
Devstral Small 1.1
Mistral AI
MA
Magistral Medium
Mistral AI
MA
Magistral Small
Mistral AI
MA
Min istral 3
Mistral AI
MA
Pixtral Large
Mistral AI
FAQ
Common questions about Mistral Large 3.
When was Mistral Large 3 released?
Mistral Large 3's default version was released on Dec 4, 2025.
How much does Mistral Large 3 cost?
Mistral Large 3's official API price is $0.50 per million input tokens and $1.5 per million output tokens via Mistral. The lowest tracked third-party offer starts at $0.50 input and $1.5 output via Merge Gateway.
Who created Mistral Large 3?
Mistral Large 3 was created by Mistral AI.
What is the context window for Mistral Large 3?
The default version has a 262.1K token context window.
Is Mistral Large 3 open weight?
Yes. The default version is marked as open weight under Apache 2.0.
How many API providers offer Mistral Large 3?
4 provider offerings are linked to the default version.
What models should I compare Mistral Large 3 with?
Nearby ranked alternatives include Claude Opus 3, Grok 2 mini, Mistral Small 3.2 24B.