MiniMax-M2.1: benchmark scores, price and ranking
MiniMax-M2.1 from MiniMax ranks #228 of 743 LLMs on the LLMs Tiger aggregate benchmark score (49.9), based on 5 benchmark results captured up to 2026-10-13.
- Organisation
- MiniMax
- Type
- Open-weight / local model
- Open weights
- Yes
- Released
- 2025-12-23 (source: epoch)
- Cheapest API price
- $0.300 in / $1.20 out per million tokens (via bedrock, observed 2026-10-13T20:47Z)
- Aggregate rank
- #228 of 743 (score 49.9)
Relative to other models, MiniMax-M2.1 is strongest on MMLU-Pro (#3 of 136) and weakest on ALE-Bench (#51 of 79).
Across the 5 benchmarks where at least ten models are scored, MiniMax-M2.1 is in the top 10% on 1 and in the top half on 2.
Its cheapest listed input price, $0.300 per million tokens, is lower than that of 58% of the 427 models with a listed price.
MiniMax-M2.1 benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| MMLU-Pro | 88.0% | #3 of 136 | HuggingFace MMLU-Pro leaderboard | 2026-10-13 |
| GPQA Diamond | 80.8% | #101 of 275 | HuggingFace GPQA Diamond leaderboard | 2026-10-13 |
| Terminal-Bench 2.0 | 36.6% | #15 of 28 | Epoch AI Benchmarking Hub | 2026-10-13 |
| WebDev Arena | 1,387 | #39 of 67 | Epoch AI Benchmarking Hub | 2026-10-13 |
| ALE-Bench | 624 | #51 of 79 | Epoch AI Benchmarking Hub | 2026-10-13 |
About MiniMax-M2.1
MiniMax-M2.1 is an open-weight model that can run on your own hardware from MiniMax. It was released on 2025-12-23, according to Epoch AI. The cheapest API listing we track charges $0.300 per million input tokens and $1.20 per million output tokens, through bedrock. We hold 5 benchmark results for it across 2 areas, and 1 is in the top tenth of its board.
Where MiniMax-M2.1 stands
On our aggregate score, MiniMax-M2.1 is #228 of 743 ranked models, around the middle of the table. Its score is 49.9; the scale is centred on 50, so it sits 0.1 points below the typical model. That beats 69% of the ranked field. Among the 6 ranked MiniMax models it is #2. The next one up is MiniMax-M1-80k at 50.9. The next one down is MiniMax-M2 at 48.7.
MiniMax-M2.1, benchmark by benchmark
On MMLU-Pro, MiniMax-M2.1 scores 88.0% and ranks #3 of 136. Only 2 models score higher, led by Darwin-180B-RSI at 88.1%, which is 0.1 points ahead. It covers 57-subject knowledge + reasoning. The board median is 55.3%, so MiniMax-M2.1 is 32.7 points above it. Source: HuggingFace MMLU-Pro leaderboard, captured 2026-10-13.
On GPQA Diamond, MiniMax-M2.1 reaches 80.8%. It ranks #101 of 275, ahead of 63% of the field, and 15.0 points behind GPT-6 Astra (max) at 95.8%. The results either side are GLM-4.6-FP8 (81.3%) and EXAONE-4.5-33B (80.5%). It covers expert-level science questions. The board median is 71.7%, so MiniMax-M2.1 is 9.1 points above it. Source: HuggingFace GPQA Diamond leaderboard, captured 2026-10-13.
MiniMax-M2.1 lands below the middle on Terminal-Bench 2.0: 36.6%, #15 of 28, with 50% of the field scoring higher. Gemini 3 Pro Preview leads at 69.4%; the closest result above is Claude Opus 4.1 (unknown thinking) at 38.0%. It covers real terminal tasks (agent + model). The board median is 37.3%, so MiniMax-M2.1 is 0.7 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
MiniMax-M2.1 lands below the middle on WebDev Arena: 1,387, #39 of 67, with 57% of the field scoring higher. Claude Opus 5.5 (max) leads at 1,820; the closest result above is Claude Opus 4.1 (unknown thinking) at 1,389. It covers building web apps, crowd-voted (Elo). The board median is 1,433, so MiniMax-M2.1 is 46 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
MiniMax-M2.1 lands below the middle on ALE-Bench: 624, #51 of 79, with 63% of the field scoring higher. GPT-6 Astra (max) leads at 2,951; the closest result above is MiniMax-M3 at 640. It covers AtCoder heuristic-optimisation contests (performance rating). The board median is 764, so MiniMax-M2.1 is 140 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Strengths and weaknesses
Software engineering: uneven, ahead of 41% of the field on average across 3 benchmarks. Its best board here is Terminal-Bench 2.0 (#15 of 28) and its weakest is ALE-Bench (#51 of 79). Reasoning, maths and science: strong, ahead of 81% of the field on average across 2 benchmarks. Its best board here is MMLU-Pro (#3 of 136) and its weakest is GPQA Diamond (#101 of 275).
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so MiniMax-M2.1 costs less than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4.1 Flash (high) ($0.010, 48.9), DeepSeek V4 Flash 0731 (max) ($0.016, 53.1). Similar scores cost more on qwen-2.5-72b-instruct ($0.380) and Devstral (2512) ($0.400).
Release history
The MiniMax model released before it was MiniMax-M2 on 2025-10-27, which scores 48.7. The next one was MiniMax-M2.5 on 2026-02-12, at 46.8.
Reading these numbers
The 5 scores on this page come from 3 sources: Epoch AI Benchmarking Hub (3), HuggingFace GPQA Diamond leaderboard (1), HuggingFace MMLU-Pro leaderboard (1). No official leaderboard supplied any of them, so treat them as third-party measurements. With 5 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #225 claude-sonnet-4-20250514_32K (49.9)
- #226 Mistral-Small-4-119B-2603 (49.9)
- #227 DeepSeek Coder V2 0724 (49.9)
- #229 claude-sonnet-4-20250514_12K (49.9)
- #230 Claude Opus 4.5 (128k thinking) (49.9)
- #231 Mistral Large 2 (2407) (49.9)
More from MiniMax
- MiniMax-M1-80k (#160, 50.9)
- MiniMax-M2 (#330, 48.7)
- MiniMax-M2.5 (#484, 46.8)
- MiniMax-M3 (#566, 45.6)
- MiniMax-M2.7 (#568, 45.6)
Compare it on the live leaderboard → · How the aggregate score works