Contamination Map Methodology

Text / LLM / Model

Mistral Small 4

Organisation
Mistral AI
Family
Mistral · version 2603
Released
2026-03-16 (Mistral AI blog 'Introducing Mistral Small 4' dated March 16, 2026 (https://mistral.ai/news/mistral-small-4))
Training cutoff
2024-11-01
Evidence
“Your knowledge base was last updated on Friday, November 1, 2024.”
huggingface.co · code · SYSTEM_PROMPT.txt shipped with the weights · retrieved 2026-10-15
Checkpoints
mistralai/Mistral-Small-4-119B-2603, mistralai/Mistral-Small-4-119B-2603-NVFP4
Weights
Open · Apache 2.0
Links
Model card

Training data disclosure: None

Level determined for 4% of benchmark cells.

Benchmark cells

Classic LLM benchmarks

Unknown MMLU (Measuring Massive Multitask Language Understanding) · training data not disclosed in enough detail

Unknown GSM8K (Grade School Math 8K) · training data not disclosed in enough detail

Unknown MATH (Mathematics Aptitude Test of Heuristics) · training data not disclosed in enough detail

Unknown HumanEval · training data not disclosed in enough detail

Unknown ARC-Challenge (AI2 Reasoning Challenge, Challenge Set) · training data not disclosed in enough detail

Unknown HellaSwag · training data not disclosed in enough detail

Unknown TruthfulQA · training data not disclosed in enough detail

Unknown WinoGrande · training data not disclosed in enough detail

Multimodal LLM benchmarks

Unknown MMMU (Massive Multi-discipline Multimodal Understanding and Reasoning) · training data not disclosed in enough detail

Unknown MMMU-Pro · training data not disclosed in enough detail

Unknown MathVista · training data not disclosed in enough detail

Unknown ChartQA · training data not disclosed in enough detail

Unknown DocVQA · training data not disclosed in enough detail

Unknown AI2D (AI2 Diagrams) · training data not disclosed in enough detail

Unknown CharXiv · training data not disclosed in enough detail

Recent LLM benchmarks

Unknown GPQA Diamond · training data not disclosed in enough detail

Unknown MMLU-Pro · training data not disclosed in enough detail

Unknown LiveCodeBench (code generation, release_v6) · training data not disclosed in enough detail

Unknown SWE-bench Verified · training data not disclosed in enough detail

Unknown AIME 2024 · training data not disclosed in enough detail

Not seen AIME 2025 · its test items were published from 2025-02-06, after the model's stated training-data cutoff (2024-11-01)

Evidence (3)
“Your knowledge base was last updated on Friday, November 1, 2024.”
huggingface.co · code · SYSTEM_PROMPT.txt shipped with the weights · retrieved 2026-10-15
“2025 AIME I problems and solutions. The test was held on Thursday, February 6, 2025.”
web.archive.org · website · retrieved 2026-10-15
“This year the AIME I will take place on February 6, 2025, while the alternate AIME II will take place on February 12, 2025.”
maa.org · website · retrieved 2026-10-15

Unknown MATH-500 · training data not disclosed in enough detail

Notes

Single checkpoint unifies instruct, reasoning (Magistral) and coding (Devstral) behaviour; no training-data description. Cutoff comes only from the recommended system prompt, not from a data statement. Accepts image input: “Accepts both text and image input, with text output.”

Sources