MiniMax, open weights
MiniMax-M2.7
MiniMax-M2.7 by MiniMax ranks 196th of 354 ranked models on the Noometry Index as of October 2026, with a score of 37.7. Its strongest category is long context, where it ranks 99th. API pricing starts at $0.30 per million input tokens and $1.20 per million output tokens, with a 205K-token context window.
Last verified
Specifications
- Noometry rank
- #196 of 354
- Index score
- 37.7
- Evidence
- Confirmed 30 results
- Provider
MiniMax- Released
- March 18, 2026
- Weights
- Open weights
- Reasoning
- Yes
- Context window
- 205K
- Max output
- 131K
- Input price
- $0.30 / M
- Output price
- $1.20 / M
- Blended price
- $0.52 / M
- Output speed
- Not measured
- Value
- #83 of 219
- Knowledge cutoff
- Unknown
- Input
- text
- Hugging Face
- MiniMaxAI/MiniMax-M2.7
Category scores
Each category score combines every public result we have in that category.
- Coding 41.8
- Agentic & Tool Use 25.1
- Reasoning 19.7
- Math 25.9
- Knowledge 37.7
- Multilingual 50.3
- Instruction Following 74.1
- Long Context 43.3
- Writing & Preference 58.9
| Category | Score | Rank | Results |
|---|---|---|---|
| Coding | 41.8 | #120 | 4 |
| Agentic & Tool Use | 25.1 | #111 | 3 |
| Reasoning | 19.7 | #253 | 4 |
| Math | 25.9 | #263 | 2 |
| Knowledge | 37.7 | #152 | 2 |
| Multilingual | 50.3 | #123 | 1 |
| Instruction Following | 74.1 | #103 | 1 |
| Long Context | 43.3 | #99 | 1 |
| Writing & Preference | 58.9 | #112 | 3 |
Strengths and weaknesses
Categories where MiniMax-M2.7 places highest and lowest among the models ranked in each, with its score against that category's median.
Strongest categories
| Category | Score | vs median | Rank |
|---|---|---|---|
| Long Context | 43.3 | +2.4 | #99 of 296, top 34% |
| Instruction Following | 74.1 | +2.8 | #103 of 305, top 34% |
| Coding | 41.8 | +3.0 | #120 of 340, top 36% |
Weakest categories
| Category | Score | vs median | Rank |
|---|---|---|---|
| Math | 25.9 | −10.6 | #263 of 327, top 81% |
| Reasoning | 19.7 | −3.9 | #253 of 350, top 73% |
| Agentic & Tool Use | 25.1 | −5.3 | #111 of 154, top 73% |
Closest competitors
The models ranked just above and below MiniMax-M2.7. When scores are this close, price and speed are often the better way to choose.
| Model | Rank | Score | Blended $/M | Speed | |
|---|---|---|---|---|---|
| GPT-5-Codex | #192 | 37.9 | $3.44 | 95 | Compare |
| Hunyuan Standard 2025 02 10 | #193 | 37.9 | — | — | Compare |
| Gemini 2.0 Flash-Lite | #194 | 37.8 | — | — | Compare |
| DeepSeek-R1-Distill-Llama-70B | #195 | 37.8 | — | 18 | Compare |
| Gemini Advanced 0514 | #197 | 37.7 | — | — | Compare |
| Grok 2 Mini 2024 08 13 | #198 | 37.7 | — | — | Compare |
| Mercury | #199 | 37.6 | — | 35 | Compare |
| DeepSeek-V2.5 (Sep 2024) | #200 | 37.6 | — | — | Compare |
Sponsored placements are available on pages like this one. Advertise on Noometry
Benchmark results
Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.
Coding
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena WebDev | 1398 | #73 of 113, top 65% | LMArena | 2026-10-08 | |
| SciCode | 47% | #54 of 121, top 45% | Epoch AI | ||
| WeirdML | 37% | #88 of 119, top 74% | Epoch AI | ||
| LMArena Coding | 1454 | #78 of 294, top 27% | LMArena | 2026-10-08 | |
| ALE-Bench | 599.25 | #80 of 105, top 77% | Epoch AI |
Agentic & Tool Use
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| Terminal-Bench | 45.1% | #20 of 41, top 49% | Epoch AI | ||
| ExploitBench | 13.3% | #9 of 9, top 100% | Epoch AI | ||
| GBAEval | 0% | #23 of 23, top 100% | Epoch AI |
Reasoning
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| NYT Connections (extended) | 24.7% | #77 of 91, top 85% | Lech Mazur benchmarks | ||
| CritPt | 0.6% | #89 of 134, top 67% | Epoch AI | ||
| Thematic Generalization | 39.3% | #21 of 23, top 92% | Lech Mazur benchmarks | ||
| LMArena Hard Prompts | 1422 | #101 of 297, top 35% | LMArena | 2026-10-08 | |
| Epoch Capabilities Index | 145.85 | #77 of 213, top 37% | Epoch AI | 2026-03-18 |
Math
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| ProofBench | 3% | #73 of 77, top 95% | Epoch AI | ||
| LMArena Math | 1420 | #100 of 285, top 36% | LMArena | 2026-10-08 |
Knowledge
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| Vectara Hallucination Rate (lower is better) | 12.9% | #80 of 96, top 84% | Vectara Hallucination Leaderboard | ||
| LMArena Expert | 1444 | #78 of 273, top 29% | LMArena | 2026-10-08 |
Multilingual
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Non-English | 1382 | #123 of 297, top 42% | LMArena | 2026-10-08 | |
| LMArena Chinese | 1441 | #112 of 285, top 40% | LMArena | 2026-10-08 | |
| LMArena French | 1421 | #99 of 223, top 45% | LMArena | 2026-10-08 | |
| LMArena German | 1398 | #95 of 231, top 42% | LMArena | 2026-10-08 | |
| LMArena Japanese | 1262 | #136 of 211, top 65% | LMArena | 2026-10-08 | |
| LMArena Korean | 1313 | #122 of 213, top 58% | LMArena | 2026-10-08 | |
| LMArena Russian | 1383 | #127 of 283, top 45% | LMArena | 2026-10-08 | |
| LMArena Spanish | 1403 | #108 of 226, top 48% | LMArena | 2026-10-08 |
Instruction Following
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Instruction Following | 1405 | #96 of 298, top 33% | LMArena | 2026-10-08 |
Long Context
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Longer Query | 1419 | #94 of 291, top 33% | LMArena | 2026-10-08 |
Writing & Preference
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Text | 1405 | #116 of 297, top 40% | LMArena | 2026-10-08 | |
| LMArena Creative Writing | 1354 | #126 of 295, top 43% | LMArena | 2026-10-08 | |
| LMArena Multi-Turn | 1412 | #106 of 295, top 36% | LMArena | 2026-10-08 |
API pricing by provider
| Route | Input $/M | Output $/M | Cached input $/M | Checked |
|---|---|---|---|---|
| deepinfra | $0.25 | $1 | $0.05 | 2026-10-10 |
| minimax | $0.30 | $1.20 | $0.06 | 2026-10-10 |
| openrouter | $0.21 | $0.84 | $0.042 | 2026-10-10 |
| together | $0.30 | $1.20 | $0.06 | 2026-10-10 |
Compare MiniMax-M2.7
- MiniMax-M2.7 vs MiniMax-M2.5
- MiniMax-M2.7 vs DeepSeek-R1-Distill-Llama-70B
- MiniMax-M2.7 vs Gemini Advanced 0514
- MiniMax-M2.7 vs Gemini 2.0 Flash-Lite
- MiniMax-M2.7 vs Grok 2 Mini 2024 08 13
- MiniMax-M2.7 vs Hunyuan Standard 2025 02 10
- MiniMax-M2.7 vs Mercury
- MiniMax-M2.7 vs GPT-6 Astra
- MiniMax-M2.7 vs Claude Fable 5.1
- MiniMax-M2.7 vs Gemini 3.8 Flash
- MiniMax-M2.7 vs Kimi K3
- MiniMax-M2.7 vs Grok 4.6
- MiniMax-M2.7 vs Qwen3.8 Max
- MiniMax-M2.7 vs GLM-5.3
Other MiniMax models
Frequently asked questions
How good is MiniMax-M2.7?
MiniMax-M2.7 by MiniMax ranks 196th of 354 ranked models on the Noometry Index as of October 2026, with a score of 37.7. Its strongest category is long context, where it ranks 99th. API pricing starts at $0.30 per million input tokens and $1.20 per million output tokens, with a 205K-token context window.
How much does MiniMax-M2.7 cost?
MiniMax-M2.7 costs $0.30 per million input tokens and $1.20 per million output tokens on MiniMax's own API, with cached input at $0.06.
What is MiniMax-M2.7's context window?
MiniMax-M2.7 accepts up to 205K tokens of input and can write up to 131K tokens in one response.
Is MiniMax-M2.7 open source?
Yes. MiniMax-M2.7's weights are downloadable from Hugging Face (MiniMaxAI/MiniMax-M2.7); check the license for commercial terms.
What are MiniMax-M2.7's strengths and weaknesses?
Relative to other ranked models, MiniMax-M2.7 places best in long context, instruction following, coding and lowest in math, reasoning, agentic & tool use.
What is MiniMax-M2.7 best at?
Its best category is long context, where it ranks 99th on Noometry.