Mistral Large 2407
Jul 2024
Retired 2025-03-30Replace withMistral Large 4Newer version
MiMo v2.6 ProCheaper, as good
GPT-6 LunaBetter
Qwen3.8 FlashOpen weights
Mistral Large 4Newer versionMiMo v2.6 ProCheaper, as goodGPT-6 LunaBetterQwen3.8 FlashOpen weightsMistral Large 2407Compare inon
Benchmarks of Mistral Large 240724
BenchmarkByScoreRank
Overall1
Knowledge13
Math1
Science1
Documents1
Legacy · saturated or no longer reported7
No benchmark matches
Rankingin non-thinking mode?
Retired models are left out of the current rankings.
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperMuse Spark 1.3max×7.1-33%$1.60Only auto tool_choice is supported+5BetterGPT-6 Lunalow×3.2-93%$0.005Function calling only with none · Chat Completions+3OpenQwen3.8 Flashthinking×5.9-92%$0.37Memory (4-bit): 74 GB → 108 GB+6CheaperMiMo v2.6 ProthinkingOpen×6.8-82%$0.13Memory (4-bit): 74 GB → 615 GB+4
CheaperDeepSeek V4.1 Flash 1017maxOpen×6.6-83%$0.26Memory (4-bit): 74 GB → 458 GB+5ModelScoreCostPer taskWhat changes
Muse Spark 1.3max×7.1-33%$1.60Only auto tool_choice is supported+5MiMo v2.6 ProthinkingOpen×6.8-82%$0.13Memory (4-bit): 74 GB → 615 GB+4DeepSeek V4.1 Flash 1017maxOpen×6.6-83%$0.26Memory (4-bit): 74 GB → 458 GB+5
GLM 5.3maxOpen×6.6-28%$2.01Memory (4-bit): 74 GB → 452 GB+4
Step 5 PreviewthinkingOpen×6.4-52%$1.03Memory (4-bit): 74 GB → 363 GB+3ModelScoreCostPer taskWhat changes
GPT-6 Lunalow×3.2-93%$0.005Function calling only with none · Chat Completions+3
Granite 4.2 3BthinkingOpen+33%-98%$0.006Context: 131K → 128K
Gemma 4 E4BthinkingOpen+31%-99%$0.008Context: 131K → 128KGPT-5.6 Lunalow×3.1-85%$0.01GPT OSS 20BhighOpen+32%-99%$0.011ModelScoreCostPer taskWhat changes
Qwen3.8 FlashthinkingOpen×5.9-92%$0.37Memory (4-bit): 74 GB → 108 GB+6Qwen3.8 27BxhighOpen×5-68%$1.01max_tokens range · QwenCloud+4
Nemotron 3.5 Super VLOpen×3.9-94%$0.19Memory (4-bit): 74 GB → 75 GBTInkling SmallthinkingOpen×3.8-79%$0.089Memory (4-bit): 74 GB → 160 GB+1IK2 Horizon MoVA 36B A4BthinkingOpen×3.7-42%$1 / $4ModelScoreCostPer taskWhat changes
Mistral Large 4thinking×5.6—$1.13Infinite whitespace stream+4Mistral Medium 3.5highOpen×2.1—$0.50Memory (4-bit): 74 GB → 77 GBMistral Small 4thinkingOpen+66%—$0.015Mistral Large 3non-thinkingOpen+36%—$0.031Memory (4-bit): not verifiedMagistral Medium 1thinking+33%——Context: 131K → 40KDevstral Mediumnon-thinkingOpen+32%—$0.40 / $2Context: 131K → 128K+1Magistral Small 1thinkingOpen+21%—$0.70 / $1.16Context: 131K → 128KMistral Small 3non-thinkingOpensame—$0.05 / $0.08Context: 131K → 32KAll Mistral models →Run it yourself
estimateFits on✗ 8 GB (phone, laptop)✗ 16 GB✗ 24 GB GPU✗ 48 GB GPU✓ 80 GB GPU · compressed
Licencemrlcommercial use with conditions
History
4 versionsFAQ
How much does Mistral Large 2407 cost?›
$2 per 1M input tokens and $6 per 1M output tokens; cached input $0.20 per 1M.
Is Mistral Large 2407 free to use?›
There is no free API; the cheapest paid price is $2 per 1M input tokens. The weights are open, so you can run it yourself.
How can I use Mistral Large 2407?›
Through the API of AWS Bedrock, Mistral.
What is the context window of Mistral Large 2407?›
131K tokens, with up to 105K output tokens.
When was Mistral Large 2407 released? Is there a newer version?›
Released 2024-07-24. It has been superseded by Mistral Large 3.
Is Mistral Large 2407 open source?›
Yes, the weights are open (mrl licence).
Can I run Mistral Large 2407 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Mistral Large 2407 read images?›
It accepts text and PDF files as input.
What is Mistral Large 2407?›
Mistral Large 2407 is an open-weights language model from Mistral, released Jul 2024, the successor to Mistral Large.
What are the best alternatives to Mistral Large 2407?›
Muse Spark 1.3 (cheaper), GPT-6 Luna (better), Qwen3.8 Flash (open weights).