[whatmodel.sh](../../README.md) / Mistral Large 3 (675B-A41B)
# Mistral Large 3 (675B-A41B)
Mistral · released 2025-12 · 675B params (41B active) · apache-2.0
**score 61.5** ±1.2 · #189 of 732 · from 15 benchmark results across 7 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks Mistral Large 3 (675B-A41B) among every model on that task or test, worst to best.
task vs the others rank counts tests
agent tasks ▁▁▃▆●▄▅▅▄▄▂▁ #222 of 313* 3.8% 1
coding ▁▁▁▃▅▇█●▅▄▃▁ #181 of 469* 27% 1
chat & writing ▁▁▁▃▆▄▆█▆●▇▃ #101 of 402* 20% 2
vision & documents ▁▁▂▂▁▃▄▂●▇█▄ #87 of 153* 15% 2
agent tasks · #222 of 313 · 3.8% of its score
Beats [o1 (2024-12-17)](o1-2024-12-17.md), [o3-mini](o3-mini.md), [DeepSeek R1 (Jan '25)](deepseek-r1.md) here, though they rank higher overall. Trails [SmolLM3 3B](smollm3-3b.md), [Jamba 1.5 Mini](jamba-1-5-mini.md), [Mellum2 12B-A2.5B Instruct](mellum2-12b-a2-5b-instruct.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
Thinkingbox-bench pass@1 (507 sta… 4.7% ●··▃·▃·█▃▅██ beats 0/15 very hard · avg 5.1% 3.8% sharp
coding · #181 of 469 · 27% of its score
Beats [Qwen3.5 9B](qwen3-5-9b.md), [Gemini 2.5 Flash](gemini-2-5-flash.md), [Qwen3-VL 32B Thinking](qwen3-vl-32b-thinking.md) here, though they rank higher overall. Trails [QwQ-32B-Preview](qwq-32b-preview.md), [Ministral 3 8B Reasoning 2512](ministral-3-8b-reasoning.md), [Phi-4-reasoning 14B](phi-4-reasoning.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
Arena WebDev (Code Arena) 1232.6 ▁▁●▂▅█▃▆▅▅▁▂ beats 8/129 avg 981 27% sharp
chat & writing · #101 of 402 · 20% of its score
Beats [LongCat 2.0 (1.6T-A48B)](longcat-2-0.md), [Inkling Small (266B-A12B)](inkling-small.md), [GPT-5.4 nano](gpt-5-4-nano.md) here, though they rank higher overall.
test result vs the others in its run difficulty counts why
LMArena (text) 1414.3 ▁▁▁▂▃▃▃▆▆●█▃ beats 263/385 avg 1260 15% blunt
Arena-Hard (version/judge unspeci… 66.7% ▃▃▂▅▃▆▅●▅▅▂▃ beats 17/30 medium · avg 47% 4.5% at its level · small run
vision & documents · #87 of 153 · 15% of its score
Beats [DeepSeek V4 Flash Vision (Exp)](deepseek-v4-flash-vision-exp.md), [GPT-5 mini](gpt-5-mini.md), [o1 (2024-12-17)](o1-2024-12-17.md) here, though they rank higher overall. Trails [GPT-4.1 mini](gpt-4-1-mini.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
Arena vision (style controlled) 1199.1 ▁▁▂▂▂▂▃▄●▅█▇ beats 72/152 avg 1061 10% blunt
MMMU Pro (maker) 54.0% ▂▃▁▃▆●▃▇▇██▅ beats 16/52 hard · avg 41% 4.8% at its level · small run
other tests (9)
test result vs the others in its run difficulty counts why
GPQA Diamond (maker) 64.1% ▁▁▁▁▂▄▅●███▄ beats 129/322 hard · avg 42% 9.4% sharp · at its level · small run
MMLU-Pro (maker) 80.9% ▁▁·▁▁▁▃▂▄▅●▁ beats 130/188 medium · avg 59% 5.4% small run
LongSocialBench (full-context, 4-… 42.0% ▆▂·▄●·▂▂·▄▄▂ beats 9/16 very hard · avg 36% 4.9% blunt
Icelandic LLM Leaderboard average… 25.7% ▁▁▃▂▂●▁▃▄▄▄█ beats 9/38 very hard · avg 14% 4.6% at its level
IFBench (aggregation unspecified)… 37.8% ▂▆▂▄●▄▂▄▄▆█▇ beats 25/86 hard · avg 29% 4.1% at its level · small run
WMT25 Terminology Translation (te… 91.2% ▂·▂··▂·▄▂·█● beats 11/12 medium · avg 63% 3.1% noisy
WMT24++ 78.3% █·▃·●▅▃···▃▃ beats 4/10 easy · avg 70% 1.4% blunt
WMT26 General MT (GEMBA-ESA, LLM … 81.6% ▄·▄▄▄·▄██▄●▄ beats 11/12 medium · avg 65% 1.2% blunt · at its level
Vectara hallucination leaderboard… 85.5% ▁▂▁●▅█▂▃▄▃▁▁ beats 12/98 very easy · avg 90% 0.0% blunt
source results share of the score
LMArena leaderboard dataset 3 52% ▇▇▇▇▇▇
mistralai/Mistral-Small-4-119B-2603 mod… 5 28% ▇▇▇
Cohere 3 6% ▇
LongSocialBench: Do Long-Context LLMs U… 1 5% ▇
Miðeind Icelandic LLM Leaderboard 1 5% ▇
One Success Isn't Reliability: Thinking… 1 4% ▇
Vectara hallucination leaderboard 1 0% ▇
prior: pull toward the middle 1% ▇
weighted average = 61.5
Licences: LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · mistralai/Mistral-Small-4-119B-2603 model card ([Apache-2.0](../../licenses/cards-gap-NOTICE.txt)) · Cohere ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · LongSocialBench: Do Long-Context LLMs Understand Online Discussion Threads? ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Miðeind Icelandic LLM Leaderboard ([CC BY 4.0](../../licenses/mideind-icelandic-llm-leaderboard-CC-BY-4.0.txt)) · One Success Isn't Reliability: Thinkingbox ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Vectara hallucination leaderboard ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0))
## How to run it
### Through an API
api (Mistral AI) · $0.75/M blended · $0.5 in / $1.5 out per 1M tokens · [docs.mistral.ai](https://docs.mistral.ai/models/mistral-large-3-25-12)
[price source](https://github.com/BerriAI/litellm/blob/c19ce71bcdad234ca87a7af10cee45375ce811a8/model_prices_and_context_window.json): USD per 1M text tokens, standard uncached input and output. Product: La Plateforme chat completions. API model: mistral-large-2512. Settings: service tier=standard, batch=no, cache=uncached input. Additional/excluded charges: Tools, cache storage, priority service and other modalities excluded.. Currency: USD. · [context source](https://github.com/BerriAI/litellm/blob/c19ce71bcdad234ca87a7af10cee45375ce811a8/model_prices_and_context_window.json): provider maximum input tokens; not an intrinsic model maximum or legacy max\_tokens
### llama.cpp (recommended)
1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)
```sh
# macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
# Windows
irm https://llama.app/install.ps1 | iex
```
2. Run Mistral Large 3 (675B-A41B) (ud-q4_k_xl, needs ~405.5 GB, from [huggingface.co/unsloth](https://huggingface.co/unsloth/Mistral-Large-3-675B-Instruct-2512-GGUF/tree/main))
```sh
llama cli -hf unsloth/Mistral-Large-3-675B-Instruct-2512-GGUF:UD-Q4_K_XL
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf unsloth/Mistral-Large-3-675B-Instruct-2512-GGUF:UD-Q4_K_XL
```
Other versions: ud-q2_k_xl (~203 GB) `llama cli -hf unsloth/Mistral-Large-3-675B-Instruct-2512-GGUF:UD-Q2_K_XL` · ud-iq1_s (~101.75 GB) `llama cli -hf unsloth/Mistral-Large-3-675B-Instruct-2512-GGUF:UD-IQ1_S`
### LM Studio
1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download)
2. Run Mistral Large 3 (675B-A41B) (ud-q4_k_xl, needs ~405.5 GB, from [huggingface.co/unsloth](https://huggingface.co/unsloth/Mistral-Large-3-675B-Instruct-2512-GGUF/tree/main))
```sh
lms get https://huggingface.co/unsloth/Mistral-Large-3-675B-Instruct-2512-GGUF@UD-Q4_K_XL -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: ud-q2_k_xl (~203 GB) `lms get https://huggingface.co/unsloth/Mistral-Large-3-675B-Instruct-2512-GGUF@UD-Q2_K_XL -y` · ud-iq1_s (~101.75 GB) `lms get https://huggingface.co/unsloth/Mistral-Large-3-675B-Instruct-2512-GGUF@UD-IQ1_S -y`
Model repository and documentation: [huggingface.co/mistralai](https://huggingface.co/mistralai/Mistral-Large-3-675B-Instruct-2512)
[interactive version](https://whatmodel.sh/model/mistral-large-3) · [JSON](https://whatmodel.sh/api?format=json&all=1)