Magistral Small 1.2: benchmark scores, price and ranking
Magistral Small 1.2 from Mistral AI ranks #663 of 743 LLMs on the LLMs Tiger aggregate benchmark score (43.8), based on 7 benchmark results captured up to 2026-10-13.
- Organisation
- Mistral AI
- Type
- Open-weight / local model
- Open weights
- Yes
- Released
- 2025-09-18 (source: epoch)
- Aggregate rank
- #663 of 743 (score 43.8)
Relative to other models, Magistral Small 1.2 is strongest on Epoch ECI (#85 of 204) and weakest on SciCode (#83 of 106).
Across the 7 benchmarks where at least ten models are scored, Magistral Small 1.2 is in the top 10% on 0 and in the top half on 1.
Magistral Small 1.2 benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| Epoch ECI | 131 | #85 of 204 | Epoch AI Benchmarking Hub | 2026-10-13 |
| OTIS Mock AIME | 28.1% | #125 of 189 | Epoch AI Benchmarking Hub | 2026-10-13 |
| DTBench | 61.3% | #99 of 146 | Epoch AI Benchmarking Hub | 2026-10-13 |
| CritPt | 0.3% | #79 of 115 | Epoch AI Benchmarking Hub | 2026-10-13 |
| GPQA Diamond | 47.6% | #196 of 275 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Chess Puzzles | 3.0% | #90 of 122 | Epoch AI Benchmarking Hub | 2026-10-13 |
| SciCode | 35.2% | #83 of 106 | Epoch AI Benchmarking Hub | 2026-10-13 |
About Magistral Small 1.2
Magistral Small 1.2 is an open-weight model that can run on your own hardware from Mistral AI. It was released on 2025-09-18, according to Epoch AI. No API price is listed for it yet. We hold 7 benchmark results for it across 4 areas, and none is in the top tenth of its board.
Where Magistral Small 1.2 stands
On our aggregate score, Magistral Small 1.2 is #663 of 743 ranked models, in the lower half. Its score is 43.8; the scale is centred on 50, so it sits 6.2 points below the typical model. That beats 11% of the ranked field. Among the 44 ranked Mistral AI models it is #32. The next one up is Mistral Small 3.2 at 44.0. The next one down is mistral-medium-2508 at 43.7.
Magistral Small 1.2, benchmark by benchmark
Epoch ECI puts Magistral Small 1.2 at 131, #85 of 204. That beats 58% of the models tested, but 25 points short of Qwen 3.8 Max (156). The results either side are Gemini 1.5 Pro (Sept 2024) (132) and Grok-2 (Dec 2024) (130). It covers Epoch Capabilities Index: Epoch's own aggregate of ~40 benchmarks. The board median is 127, so Magistral Small 1.2 is 4 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Magistral Small 1.2 lands below the middle on OTIS Mock AIME: 28.1%, #125 of 189, with 66% of the field scoring higher. Claude Fable 5.1 (max) leads at 100.0%; the closest result above is GPT-4.1 nano at 28.9%. It covers competition maths (AIME 2024-25 style). The board median is 57.8%, so Magistral Small 1.2 is 29.7 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Magistral Small 1.2 lands below the middle on DTBench: 61.3%, #99 of 146, with 67% of the field scoring higher. Claude Opus 5.5 (max) leads at 98.9%; the closest result above is c4ai-command-a-03-2025 at 61.3%. It covers dynamic tool-use benchmark. The board median is 68.8%, so Magistral Small 1.2 is 7.5 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Magistral Small 1.2 lands below the middle on CritPt: 0.3%, #79 of 115, with 68% of the field scoring higher. GPT-5.6 Sol (max) leads at 32.3%; the closest result above is Magistral Medium 1.2 at 0.3%. It covers unpublished research-level physics problems. The board median is 2.9%, so Magistral Small 1.2 is 2.6 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Magistral Small 1.2 lands below the middle on GPQA Diamond: 47.6%, #196 of 275, with 71% of the field scoring higher. GPT-6 Astra (max) leads at 95.8%; the closest result above is gemma-3-27b-it at 47.7%. It covers expert-level science questions. The board median is 71.7%, so Magistral Small 1.2 is 24.1 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Magistral Small 1.2 lands below the middle on Chess Puzzles: 3.0%, #90 of 122, with 73% of the field scoring higher. GPT-6 Astra (max) leads at 72.0%; the closest result above is deepseek-r1-0528-qwen3-8b at 3.0%. It covers chess tactics puzzles. The board median is 12.5%, so Magistral Small 1.2 is 9.5 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Magistral Small 1.2 lands below the middle on SciCode: 35.2%, #83 of 106, with 77% of the field scoring higher. Claude Opus 5.5 (max) leads at 66.9%; the closest result above is Qwen3-32B at 35.4%. It covers scientific research code. The board median is 44.3%, so Magistral Small 1.2 is 9.1 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Strengths and weaknesses
Reasoning, maths and science: uneven, ahead of 31% of the field on average across 3 benchmarks. Its best board here is OTIS Mock AIME (#125 of 189) and its weakest is GPQA Diamond (#196 of 275). General knowledge and chat: uneven, ahead of 42% of the field on average across 2 benchmarks. Its best board here is Epoch ECI (#85 of 204) and its weakest is Chess Puzzles (#90 of 122).
Release history
The Mistral AI model released before it was Magistral Medium 1.2 on 2025-09-18, which scores 47.1. The next one was Mistral Large 3 on 2025-12-02, at 42.7.
Reading these numbers
The 7 scores on this page come from 1 source: Epoch AI Benchmarking Hub (7). No official leaderboard supplied any of them, so treat them as third-party measurements. With 7 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #660 gemini-1.5-pro-002 (43.9)
- #661 GPT-5 nano (high) (43.8)
- #662 gemma-2-27b-it (43.8)
- #664 GPT-4 (Mar 2023) (43.7)
- #665 gemma-2-27b (43.7)
- #666 mistral-medium-2508 (43.7)
More from Mistral AI
- Mistral Large 2 (2407) (#231, 49.9)
- codestral-2405 (#352, 48.4)
- pixtral-large-2411 (#360, 48.3)
- Codestral-22B-v0.1-Q4_K_M (#392, 48.0)
- codestral:22b-v0.1-q8_0 (#393, 48.0)
- Mistral Large 4 (high) (#411, 47.7)
Compare it on the live leaderboard → · How the aggregate score works