available
Weights
Official FP8 weights are downloadable.
Inspect this claim’s source ↗Mistral AI / FAMILY RECORD
Mistral Small 4: license, size and what is open.
A hybrid sparse model that combines instruction following, reasoning, visual understanding, and coding modes in one downloadable checkpoint.
Sizes and download
119B total / 6.5B active in Mistral’s model docs; its release article says 6B active per token (8B including embedding and output layers)
- Mistral-Small-4-119B-2603
- Total: 119B. Active: 6.5B. Source ↗
About 70.81 GB of selected documented weights · File listing ↗
Selected artifact: mistralai/Mistral-Small-4-119B-2603-NVFP4. Format: Safetensors, official NVFP4. File evidence checked 2026-10-11. Selected artifact license: apache-2.0 ↗
Official NVFP4 variant, all 13 consolidated shards. Excludes separate FP8 repository and alternate representations. Selected weight files only; supporting files and runtime memory are separate. Not a smallest-download census or proof of hardware compatibility.
Can my computer run it?
Compatibility, runtime memory and speed are untested by Bright.
Choose your computer and inspect documented candidates →Nine things to check
Source check: . Downloads, when known, carry their own check date.
Checked release: Mistral-Small-4-119B-2603
Sources checked 2026-09-19
- Mistral Small 4 release
- · Primary source
The profile describes this checkpoint, not every release carrying the family name. Developer documentation is not independent reproduction.
THE MODEL / THE TERMS
Know what you are picking up.
- Parameters
- 119B total / 6.5B active in Mistral’s model docs; its release article says 6B active per token (8B including embedding and output layers)
- Architecture
- Mixture-of-experts transformer with 128 experts and four active experts per token.
- Modalities
- text · image
- Context
- 256K tokens
- Weights license
- Apache-2.0
- Commercial use
- Permitted by Apache-2.0.
- Memory
- The FP8 repository is about 242 GB; the NVFP4 repository is about 71 GB before runtime overhead.
- Known quantizations
- FP8 Official NVFP4 checkpoint
Specifications are developer-reported. Primary release documentation ↗ · All sources and terms
NINE DIMENSIONS / NO OPENNESS SCORE
What is actually available?
Each component has its own evidence. “Not established” means this source review did not verify availability; it is not a claim that nothing exists.
available
Architecture
Architecture, expert counts, and configuration are published.
Inspect this claim’s source ↗available
Inference code
The card documents vLLM and Transformers serving.
Inspect this claim’s source ↗not established
Training code
The full code used to train the model was not confirmed.
Inspect this claim’s source ↗partial
Training recipe
The release describes capabilities and design without a full reproducible recipe.
Inspect this claim’s source ↗not established
Data information
A sufficiently detailed training-data account was not confirmed.
Inspect this claim’s source ↗not established
Training data
No sufficiently specific public artifact was confirmed in this review.
Inspect this claim’s source ↗partial
Evaluation
Vendor results are published, including machine-readable evaluation artifacts.
Inspect this claim’s source ↗available
Commercial use
Apache-2.0 permits commercial use.
Inspect this claim’s source ↗WHY THIS RELEASE MATTERS
What to explore.
- Reasoning mode
- Visual understanding
- Agentic coding
- Efficient active parameter count
WHAT THE RECORD DOES NOT ESTABLISH
Keep the limits.
- Training data is not documented in enough detail for reconstruction
- Performance claims are primarily vendor-reported
The registry records vendor evaluations but does not treat them as independent corroboration.
FOLLOW THE RECORD
Sources, not superlatives.
- Introducing Mistral Small 4 ↗ (institution)
- Mistral Small 4 model card ↗ (registry)
Source check: 2026-09-19. AI-assisted editorial review, with no named external expert or independent hardware validation claimed. Licenses and model cards can change. Review the linked terms for the specific artifact and intended use.
Understand open source vs open weights ↗