Project AURIONAI Use-Cases Research & Innovation
Mistral AI

Mistral AI

Startup, France. 49 models in the directory.

startupFrance
0deployments

Identity

Type
startup
Country label
France
HQ country
France
Founding country
France
Parent
not in the notes
Parent country
not in the notes
Country basis
the research notes
Where this comes from
Convention D1: a dual label, HQ first, when the HQ, founding or parent country differ.

Models by Class

LLM18

Models

ModelClassOpennessParametersReleased
Mistral AIMistral Large 4 (v26.10)multimodal LLMunknown1.05T6 Oct 2026Mistral AIShieldstral 1.0 3Bguardopen weights permissive3B28 Jul 2026Mistral AIMistral Medium 3.5 (v26.04)multimodal LLMopen weights restricted128B28 Apr 2026Mistral AIMistral Small 4multimodal LLMopen weights permissive119BMar 2026Mistral AIVoxtral Mini 4B Realtime 2602speech recognitionopen weights permissive4.43B21 Jan 2026Mistral AIDevstral 2 (123B, 2512)codeopen weights restricted123BDec 2025Mistral AIDevstral Small 2 (24B, 2512)codeopen weights permissive24BDec 2025Mistral AIMistral Large 3 (2512)multimodal LLMopen weights permissive675BDec 2025Mistral AIMinistral 3 14B Instruct (25.12)multimodal LLMopen weights permissive13.95B31 Oct 2025Mistral AIMistral Medium 3.1 (mistral-medium-2508)multimodal LLMproprietary–13 Aug 2025Mistral AICodestral 25.08codeproprietary–30 Jul 2025Mistral AIVoxtral Mini 3B 2507speech recognitionopen weights permissive4.68B1 Jul 2025Mistral AIDevstral Medium 1.0codeproprietary–Jul 2025Mistral AIDevstral Small 1.1codeopen weights permissive24BJul 2025Mistral AIMagistralreasoningunknown–10 Jun 2025Mistral AIMagistral Smallreasoningopen weights permissive24B10 Jun 2025Mistral AIMistral Medium 3 (mistral-medium-2505)multimodal LLMproprietary–7 May 2025Mistral AIMistral Small 3 (Mistral-Small-24B-Instruct-2501)LLMopen weights permissive24B30 Jan 2025Mistral AIVoxtral Smallspeech recognitionunknown24B2025Mistral AIPixtral Largemultimodal LLMopen weights restricted124B18 Nov 2024Mistral AIMistral Small (Sep '24)LLMopen weights restricted22B17 Sep 2024Mistral AIMistral NeMo Instruct 2407Mistral AI, NVIDIALLMopen weights permissive12.25B17 Jul 2024Mistral AIMixtral 8x22B InstructLLMopen weights permissive141B17 Apr 2024Mistral AIMistral Large (Feb '24)LLMproprietary–26 Feb 2024Mistral AIMamba-Codestral-7B-v0.1codeunknown7B2024Mistral AIMistral Medium (Dec '23)LLMproprietary–11 Dec 2023Mistral AIMixtral 8x7B v0.1LLMopen weights permissive46.7B11 Dec 2023Mistral AIMixtral 8x7B Instruct v0.1LLMopen weights permissive46.7B11 Dec 2023Mistral AICodestralcodeunknown––Mistral AIDevstralcodeunknown––Mistral AILeanstral-1.5-119B-A6BLLMunknown119B–Mistral AIMagistral Medium 1.2reasoningunknown––Mistral AIMagistral Small 1.2reasoningunknown––Mistral AIMinistral-3-3B-Base-2512LLMunknown3B–Mistral AIMinistral-3-8B-Instruct-2512LLMunknown8B–Mistral AIMistral 7BLLMunknown7B–Mistral AIMistral-7B-v0.3LLMunknown7B–Mistral AIMistral Embedembeddingproprietary––Mistral AIMistral-Large-Instruct-2407 (Mistral Large 2)LLMunknown––Mistral AIMistral Mediummultimodal LLMunknown––Mistral AIMistral Moderation 2guardproprietary––Mistral AIMistral OCR 4.1OCR and documentproprietary––Mistral AIMistral SmallLLMunknown––Mistral AIMistral-Small-24B-Base-2501LLMopen weights permissive––Mistral AIMistral-Small-3.1-24B-Base-2503LLMunknown24B–Mistral AIMixtral-8x22B-v0.1LLMopen weights permissive141B–Mistral AIPixtralmultimodal LLMunknown––Mistral AIVoxtral Mini Transcribe 2speech recognitionproprietary––Mistral AIMinistral 3B (24.10)edge LLMproprietary––

Scores

BenchmarkModelScoreWho ran itSetting
AI2DPixtral Large93.8 %vendor-reportedBBox
Arena Text leaderboardMistral Large 4 (v26.10)rank onlyindependent
Arena Text leaderboardMistral Medium 3.5 (v26.04)rank onlyindependent
Artificial Analysis Intelligence IndexMistral Large 4 (v26.10)38 indexindependentMistral Large 4 Preview
Artificial Analysis Intelligence IndexMistral Large 4 (v26.10)38 indexindependentPreview
Artificial Analysis Intelligence IndexMistral Medium 3.5 (v26.04)14 indexindependent
GPQA DiamondMistral Large 3 (2512)67.17 %vendor-reported
GPQA DiamondMistral Small 471.2 %vendor-reported
GSM8KMixtral 8x22B Instruct90.8 %vendor-reportedmaj@8
HarmBenchShieldstral 1.0 3B99.4 F1vendor-reported
Hugging Face Open ASR LeaderboardVoxtral Small6.62 WERmaintainer-runaverage WER
MATHMixtral 8x22B Instruct44.6 %vendor-reportedmaj@4
MMMUPixtral Large64 %vendor-reportedCoT
MT-BenchMistral Medium (Dec '23)8.6 scorevendor-reported
PolyGuardShieldstral 1.0 3B84.6 F1vendor-reported
SEA-HELMMistral Medium 3.5 (v26.04)64.71 scoremaintainer-run
SEA-HELMMistral Medium 3.5 (v26.04)70.09 scoremaintainer-run
SEA-HELMMistral Small 452.07 scoremaintainer-runCheckpoint Mistral-Small-4-119B-2603 as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SEA-HELMMistral Small 463.14 scoremaintainer-runCheckpoint Mistral-Small-4-119B-2603 as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SWE-bench MultilingualDevstral 2 (123B, 2512)61.3 %vendor-reported
SWE-bench MultilingualDevstral Small 2 (24B, 2512)55.7 %vendor-reported
SWE-bench VerifiedDevstral 2 (123B, 2512)72.2 %vendor-reported
SWE-bench VerifiedDevstral Medium 1.061.6 %vendor-reported
SWE-bench VerifiedDevstral Small 1.153.6 %vendor-reported
SWE-bench VerifiedDevstral Small 2 (24B, 2512)68 %vendor-reported
SWE-bench VerifiedMistral Medium 3.5 (v26.04)77.6 %vendor-reported
Terminal-BenchDevstral 2 (123B, 2512)32.6 %vendor-reported
Terminal-BenchDevstral Small 2 (24B, 2512)22.5 %vendor-reported
VLGuardShieldstral 1.0 3B97.7 F1vendor-reported
WildGuardTestShieldstral 1.0 3B88.1 F1vendor-reportedprompt
XSTestShieldstral 1.0 3B94.6 F1vendor-reportedrefusal
Where this comes from
Every recorded score of this organisation's models, with who ran it. Scores from different settings are not directly comparable.

Sources

  1. Models Overview docs.mistral.ai https://docs.mistral.ai/getting-started/models/models_overview/
  2. Mistral Large 4 Preview - Intelligence, Performance & Price Analysis | Artificial Analysis artificialanalysis.ai https://artificialanalysis.ai/models/mistral-large-4
  3. huggingface.co/mistralai https://huggingface.co/mistralai
  4. Speech Recognition Accuracy Leaders, 2022–2026 select.keenable.ai https://select.keenable.ai/r/speech-recognition-accuracy-leaders-2022-2026/JbURQjlVooLQ6J10NkbRVw
  5. Magistral | Mistral AI mistral.ai https://mistral.ai/news/magistral
  6. mistral.ai/about https://mistral.ai/about

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Mistral AI", https://aurion.semai.id/organisations/mistral-ai/, records of October 2026.