Universe/Audio/Speech/Mistral AI_
Venture candidateAutomated record

Mistral AI_

Mistral is a French artificial intelligence company building frontier large language models, AI agents, and developer platforms. Headquartered in Paris, it combines open-weight and commercial models with inference infrastructure.

Visit website ↗
Rising signal56.4
Rising rankLonglist
Profile completeness57%
Last public activitySep 2, 2026
01

PUBLIC TRACTION

Usage, audience and developer pull

11.2MHF model downloads149.2K per model
19.2KHF followers42.5K total likes
24.5KGitHub stars3.3K forks
75Model portfolio4 datasets · 4 Spaces

WHY IT IS ON THE RADAR

Capital-efficient model reach

  • High reach per model
  • Shipping now
  • GitHub momentum
02

MODEL FOOTPRINT

What it publishes

Audio/SpeechLanguage/TextMultimodal
Most popular HF repositorymistralai/Mistral-7B-Instruct-v0.32.9M downloads · 2.8K likes ↗
Top tasks
automatic-speech-recognition (1), text-to-speech (1), audio-text-to-text (1), text-generation (2), image-text-to-text (1)
Model sizes
Unknown (27); 10-30B (20); 3-10B (19); 100B+ (7); 30-100B (2)
Base families
Mistral
Libraries
vllm, transformers, mistral-common, lmstudio, llama.ccp, llama.cpp, transformers.js
03

DEVELOPER ACTIVITY

How it ships

Original repositories16
Forked repositories12
Pushed in 90 days15
Archived share21%
Most-starred repositorymistralai/mistral-inference10.8K stars · 1.1K forks ↗
Languages
Python, Jupyter Notebook, JavaScript, MDX, Rust, Shell, TypeScript
Topics
llm, llm-inference, mistralai
Latest release
mistralai/mistral-vibe · Aug 27, 2026
04

ACTIVITY TIMELINE

Latest observed moves

  1. GitHub pushRepository activity ↗
  2. GitHub releaseRelease v2.24.5 ↗
  3. HF modelModel published or updated ↗
  4. HF SpaceSpace published or updated ↗
  5. HF datasetDataset published or updated ↗
05

THE NEOLAB STACK

Publicly evidenced tooling

Training
Transformers
Compute
Not disclosed
Hardware
Not disclosed
Tracking
Not disclosed
Bottlenecks
Licensing / rights

“Not disclosed” means no evidence was found in the sampled public metadata or top model card—not that the lab does not use the tool.

06

COMPARABLE LABS

Nearest modality peers

HKUST Audio
Signal 38.6 · 18K downloads · Not resolved stars
Pipecat
Signal 46.6 · 10.4K downloads · 21.6K stars
Kotoba Technologies
Signal 69.7 · 45.2K downloads · 274 stars
National Library of Sweden / KBLab
Signal 70.3 · 2M downloads · 96 stars
NineNineSix
Signal 75.1 · 76.2K downloads · 754 stars

INVESTOR DILIGENCE

Confirm the team and compute inputs behind reach per model; benchmark quality versus the base family.

  • HF member count is only a public-profile proxy
  • Low declared-task coverage

MISSING FROM THIS RECORD

headquartersfoundersOrLeadershipemployeeCountfundingbusinessModelproducts

Completeness: 8 of 14 tracked field groups. Unknowns are explicit so an analyst can close them.

DATA PROVENANCE

Every claim should be auditable.

Platform counters are a point-in-time public snapshot. Editorial facts are only promoted after source review.

Hugging Face organizationObserved 2026-09-03 ↗GitHub organization APIObserved 2026-09-03 ↗

NEOLAB TRACKER PRO

Know when the thesis changes.

Weekly deltas, funding and team alerts, portfolio watchlists, CSV exports and an analyst-ready API turn a static profile into a sourcing system.

See tracker plans →