HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks

Magistral-Small-2507

mistralai/Magistral-Small-2507

kind
original
quant
bf16 (16-bit)
family
mistral
modality
—
root model
mistralai/Mistral-Small-3.1-24B-Instruct-2503
provider
mistralai
license
apache-2.0
language
en, fr, de, es, pt, it, ja, ko, ru, zh, ar, fa
context
—
library
vllm
vRAM (GB)
—
size on disk
—
downloads
513
likes
104
created
—
last modified
2025-12-22T10:24:16+00:00 (2025)
huggingface
https://huggingface.co/mistralai/Magistral-Small-2507
tags
vllmsafetensorsmistralmistral-commonenfrdeesptitjakoruzharfaidmsneplrosrsvtr

Benchmark scores (4)

benchmarkcategoryscorecontextconfevidence
AIME 2024math72.03low| **Magistral Medium 1.1** | 72.03% | 60.99% | 71.46% | 59.35% |
GPQA Diamondreasoning71.46low| **Magistral Medium 1.1** | 72.03% | 60.99% | 71.46% | 59.35% |
AIME 2025math60.99low| **Magistral Medium 1.1** | 72.03% | 60.99% | 71.46% | 59.35% |
LiveCodeBenchcoding59.35low| **Magistral Medium 1.1** | 72.03% | 60.99% | 71.46% | 59.35% |

Discussions (0)

none scraped.

Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.