Mistral Large 3
Mistral Large 3Rankingin non-thinking mode?
Alternatives?
CheaperGLM 4.7 FlashthinkingOpen×4-81%$0.061 / $0.40Context: 262K → 200K+2
BetterGPT OSS 20BhighOpen×2.5-66%$0.011Context: 262K → 131K+2OpenGLM 5.2max×4+4,022%$1.28No image input+1
CheaperStep 3.7 FlashhighOpen×4-42%$0.20 / $1.15Context: 262K → 256K+1
CheaperMiniMax M2.5thinkingOpen×3.9-30%$0.30 / $1.20Context: 262K → 205K+2GLM 4.7 FlashthinkingOpen×4-81%$0.061 / $0.40Context: 262K → 200K+2Step 3.7 FlashhighOpen×4-42%$0.20 / $1.15Context: 262K → 256K+1MiniMax M2.5thinkingOpen×3.9-30%$0.30 / $1.20Context: 262K → 205K+2
MiMo v2 Prothinking×3.9-28%$0.43 / $0.87No image input+1
Hy3 previewthinkingOpen×3.8-62%$0.18 / $0.60Context: 262K → 256K+3GPT OSS 20BhighOpen×2.5-66%$0.011Context: 262K → 131K+2Ministral 3 8B 2512Open+8%-64%$0.011
Nemotron 3 Nano 30B A3BthinkingOpen+67%-46%$0.017Context: 262K → 131K+2Mistral Small 4thinkingOpen+68%-45%$0.017Ministral 3 14B 2512non-thinkingOpen+11%-38%$0.019GLM 5.2maxOpen×4+4,022%$1.28No image input+1GLM 4.7 FlashthinkingOpen×4-81%$0.061 / $0.40Context: 262K → 200K+2Step 3.7 FlashhighOpen×4-42%$0.20 / $1.15Context: 262K → 256K+1GLM 5thinkingOpen×4+107%$1 / $3.20Context: 262K → 203K+2GLM 5.1thinkingOpen×4+2,306%$0.75Context: 262K → 203K+2Mistral Medium 3.5highOpen×3.8+1,277%$0.43Mistral Small 4thinkingOpen+68%-45%$0.017Ministral 3 14B 2512non-thinkingOpen+11%-38%$0.019Ministral 3 8B 2512Open+8%-64%$0.011Magistral Small 1thinkingOpen+8%—$0.70 / $1.16Context: 262K → 128K+2Ministral 3 3B 2512Opensame-75%$0.008Context: 262K → 131KMagistral Medium 1thinking-6%——Context: 262K → 40K+1Devstral Mediumnon-thinkingOpen-19%—$0.40 / $2Context: 262K → 128K+2All Mistral models →Run it yourself
History
4 versionsBenchmarks
all categoriesFAQ
How much does Mistral Large 3 cost?›
$0.50 per 1M input tokens and $1.50 per 1M output tokens; cached input $0.05 per 1M. A typical benchmark task costs about $0.031 at default effort.
Is Mistral Large 3 free to use?›
There is no free API; the cheapest paid price is $0.50 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Mistral Large 3, and what is it best at?›
#305 of 336 in Chat & Knowledge, #77 of 134 in Coding, #106 of 135 in Agents & Tool Use, #74 of 197 in Writing & Language, #105 of 125 in Vision & Documents.
Is Mistral Large 3 better than Mistral Large 2407?›
On τ²-bench · AA, Mistral Large 3 scores 24.6 vs 33 — lower.
How can I use Mistral Large 3?›
Through the API of AWS Bedrock, Mistral.
What is the context window of Mistral Large 3?›
262K tokens, with up to 262K output tokens.
When was Mistral Large 3 released? Is there a newer version?›
Released 2025-12-02. It has been superseded by Mistral Large 4.
Is Mistral Large 3 open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Mistral Large 3 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Mistral Large 3 read images?›
It accepts text and images, PDF files as input.
What is Mistral Large 3?›
Mistral Large 3 is an open-weights language model from Mistral, released Dec 2025, the successor to Mistral Large 2407.
What are the best alternatives to Mistral Large 3?›
GLM 4.7 Flash (cheaper), GPT OSS 20B (better), GLM 5.2 (open weights).