Language modelActive

MiniMax M2.1

MiniMax

Released: December 20, 2025
Data date: September 8, 2026
Predecessor: MiniMax M2
Successor: MiniMax M2.5

Speed: 92.3 tokens/s
Output throughput
Price: $0.3 · $1.2
Input · Output / 1M tokens

MiniMax M2.1 is the direct update to MiniMax M2 and keeps its 230B MoE base with 10B active parameters and roughly 205K tokens of context. MiniMax names multilingual programming and complex work in real workflows as its target. Both weights and API access remain available.

Compared with MiniMax M2, the provider cites improvements on complex tasks and response efficiency. MiniMax M2.5 follows directly in the documented line and pushes further toward productive agent runs.

Specifications and access

Specification	Value and source
Weights	Open weightsSource
Access	API, LocalSource
Parameters	230BSource
Active parameters	10BSource
Context window	204,800 tokensSource
Release	2025-12-20Source
Parameters	230 BSource
Active parameters	10 BSource

Published benchmarks

Health

MedCode

Overall

1.MiniMax M346.29%

2.GLM-5.141.6%

3.Nemotron 3 Ultra38.62%

4.Grok 4.338.07%

5.MiniMax M2.734.44%

6.MiniMax M2.134.08%

7.MiMo-V2.5-Pro32.48%

8.Ling 3.0 Flash32.28%

9.Grok 4.20 Reasoning32.16%

10.MiMo-V2.531.9%

higher is betterAs of October 15, 2026Scale: 0% - 100%

Source: Vals AI

gradually.ai

Methodology and context

Comparison cohort: vals-medcode:1:overall
Metric: accuracy
Scale: 0-100
Participants: 10
Methodology and context: Source model ID: minimax/MiniMax-M2.1. Source model name: MiniMax M2.1. Provider reported by source: MiniMax. Evaluator: Vals AI. Dataset type: private. Leaderboard date: 2026-10-13. Temperature: 1. Top P: 0.95. Maximum output tokens: 30000. Latency in seconds: 19.548. Standard error: 1.943. Source ID: vals-ai. Evidence capture ID: vals-medcode-2026-10-15-988a6ae1e07e. Source payload hash: 988a6ae1e07e33ce031572ed4a804be11c36ee97a1dd75cbe4740fff3b534f8c. Parser version: vals-astro-v3. Imported at: 2026-10-15

Source: Vals AI

MedScribe

Overall

1.MiniMax M387.25%

2.MiMo-V2.5-Pro83.73%

3.Ling 3.0 Flash80.9%

4.MiniMax M2.180.78%

5.MiniMax M2.779.87%

6.Grok 4.374.4%

7.GLM-5.172.27%

8.MiMo-V2.572.15%

9.Grok 4.20 Reasoning63.41%

10.Nemotron 3.5 Lightning4.27%

higher is betterAs of October 15, 2026Scale: 0% - 100%

Source: Vals AI

gradually.ai

Methodology and context

Comparison cohort: vals-medscribe:1:overall
Metric: accuracy
Scale: 0-100
Participants: 10
Methodology and context: Source model ID: minimax/MiniMax-M2.1. Source model name: MiniMax M2.1. Provider reported by source: MiniMax. Evaluator: Vals AI. Dataset type: private. Leaderboard date: 2026-10-13. Temperature: 1. Top P: 0.95. Maximum output tokens: 30000. Latency in seconds: 53.161. Standard error: 1.831. Cost per task in USD: 0.005087. Source ID: vals-ai. Evidence capture ID: vals-medscribe-2026-10-15-271fc122ddc2. Source payload hash: 271fc122ddc2b0602bd6c7947f65a9a7f0d8cab94f6dc27d62a9380a9317f9c0. Parser version: vals-astro-v3. Imported at: 2026-10-15

Source: Vals AI

Measurements without matching peer values

These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.

AIME: AIME 2024 (Show measurement, test conditions, and source)

Source value: 79.167
Score: 79.167
Metric: accuracy
Unit: %
Benchmark version: 1
Category: math
Task: aime_2024
Task label: AIME 2024
Direction: Higher is better
Comparison cohort: vals-aime:1:aime_2024
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-10-15
Methodology: Source model ID: minimax/MiniMax-M2.1. Source model name: MiniMax M2.1. Provider reported by source: MiniMax. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-04-16. Temperature: 1. Top P: 0.95. Latency in seconds: 217.936. Standard error: 1.754. Cost per task in USD: 0.022637. Source ID: vals-ai. Evidence capture ID: vals-aime-2026-10-15-7d016210b39d. Source payload hash: 7d016210b39d24351bbb5d12be09a973decac461bb2486eb1f23724fed2b7e6c. Parser version: vals-astro-v3. Imported at: 2026-10-15
Comparison methodology: Temperature: 1. Top P: 0.95
Context: The source dataset marks this benchmark as archived. The result remains visible as historical evidence and is not interchangeable with an active leaderboard.

Source Vals AI

AIME: AIME 2025 (Show measurement, test conditions, and source)

Source value: 76.667
Score: 76.667
Metric: accuracy
Unit: %
Benchmark version: 1
Category: math
Task: aime_2025
Task label: AIME 2025
Direction: Higher is better
Comparison cohort: vals-aime:1:aime_2025
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-10-15
Methodology: Source model ID: minimax/MiniMax-M2.1. Source model name: MiniMax M2.1. Provider reported by source: MiniMax. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-04-16. Temperature: 1. Top P: 0.95. Latency in seconds: 257.333. Standard error: 1.89. Cost per task in USD: 0.025666. Source ID: vals-ai. Evidence capture ID: vals-aime-2026-10-15-7d016210b39d. Source payload hash: 7d016210b39d24351bbb5d12be09a973decac461bb2486eb1f23724fed2b7e6c. Parser version: vals-astro-v3. Imported at: 2026-10-15
Comparison methodology: Temperature: 1. Top P: 0.95
Context: The source dataset marks this benchmark as archived. The result remains visible as historical evidence and is not interchangeable with an active leaderboard.

Source Vals AI

AIME: Overall (Show measurement, test conditions, and source)

Source value: 77.917
Score: 77.917
Metric: accuracy
Unit: %
Benchmark version: 1
Category: math
Task: overall
Task label: Overall
Direction: Higher is better
Comparison cohort: vals-aime:1:overall
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-10-15
Methodology: Source model ID: minimax/MiniMax-M2.1. Source model name: MiniMax M2.1. Provider reported by source: MiniMax. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-04-16. Temperature: 1. Top P: 0.95. Latency in seconds: 237.635. Standard error: 1.289. Cost per task in USD: 0.024152. Source ID: vals-ai. Evidence capture ID: vals-aime-2026-10-15-7d016210b39d. Source payload hash: 7d016210b39d24351bbb5d12be09a973decac461bb2486eb1f23724fed2b7e6c. Parser version: vals-astro-v3. Imported at: 2026-10-15
Comparison methodology: Temperature: 1. Top P: 0.95
Context: The source dataset marks this benchmark as archived. The result remains visible as historical evidence and is not interchangeable with an active leaderboard.

Source Vals AI

CaseLaw v2: Overall (Show measurement, test conditions, and source)

Source value: 55.842
Score: 55.842
Metric: accuracy
Unit: %
Benchmark version: 2
Category: legal
Task: overall
Task label: Overall
Direction: Higher is better
Comparison cohort: vals-case-law-v2:2:overall
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-10-15
Methodology: Source model ID: minimax/MiniMax-M2.1. Source model name: MiniMax M2.1. Provider reported by source: MiniMax. Evaluator: Vals AI. Dataset type: private. Leaderboard date: 2026-05-04. Temperature: 1. Top P: 0.95. Maximum output tokens: 131000. Latency in seconds: 37.357. Standard error: 0.467. Cost per task in USD: 0.008008. Source ID: vals-ai. Evidence capture ID: vals-case_law_v2-2026-10-15-293f388f7e0f. Source payload hash: 293f388f7e0fe563287d4cd19beaa7ef9821b32795dace34bd0197ed47a592d4. Parser version: vals-astro-v3. Imported at: 2026-10-15
Comparison methodology: Temperature: 1. Top P: 0.95. Maximum output tokens: 131000
Context: The source dataset marks this benchmark as archived. The result remains visible as historical evidence and is not interchangeable with an active leaderboard.

Source Vals AI

CorpFin v2: Exact Pages (Show measurement, test conditions, and source)

Source value: 61.305
Score: 61.305
Metric: accuracy
Unit: %
Benchmark version: 2
Category: finance
Task: exact_pages
Task label: Exact Pages
Direction: Higher is better
Comparison cohort: vals-corp-fin-v2:2:exact_pages
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-10-15
Methodology: Source model ID: minimax/MiniMax-M2.1. Source model name: MiniMax M2.1. Provider reported by source: MiniMax. Evaluator: Vals AI. Dataset type: private. Leaderboard date: 2026-08-12. Temperature: 1. Top P: 0.95. Maximum output tokens: 131000. Latency in seconds: 15.39. Standard error: 1.663. Cost per task in USD: 0.001629. Source ID: vals-ai. Evidence capture ID: vals-corp_fin_v2-2026-10-15-7cf2217ab372. Source payload hash: 7cf2217ab3728d5a84bffce82aab9fc52d7f810b0cbe8c65b387df47f3d31c9b. Parser version: vals-astro-v3. Imported at: 2026-10-15
Comparison methodology: Temperature: 1. Top P: 0.95. Maximum output tokens: 131000

Source Vals AI

CorpFin v2: Max Fitting Context (Show measurement, test conditions, and source)

Source value: 54.312
Score: 54.312
Metric: accuracy
Unit: %
Benchmark version: 2
Category: finance
Task: max_fitting_context
Task label: Max Fitting Context
Direction: Higher is better
Comparison cohort: vals-corp-fin-v2:2:max_fitting_context
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-10-15
Methodology: Source model ID: minimax/MiniMax-M2.1. Source model name: MiniMax M2.1. Provider reported by source: MiniMax. Evaluator: Vals AI. Dataset type: private. Leaderboard date: 2026-08-12. Temperature: 1. Top P: 0.95. Maximum output tokens: 131000. Latency in seconds: 28.337. Standard error: 1.701. Cost per task in USD: 0.041668. Source ID: vals-ai. Evidence capture ID: vals-corp_fin_v2-2026-10-15-7cf2217ab372. Source payload hash: 7cf2217ab3728d5a84bffce82aab9fc52d7f810b0cbe8c65b387df47f3d31c9b. Parser version: vals-astro-v3. Imported at: 2026-10-15
Comparison methodology: Temperature: 1. Top P: 0.95. Maximum output tokens: 131000

Source Vals AI

CorpFin v2: Overall (Show measurement, test conditions, and source)

Source value: 58.896
Score: 58.896
Metric: accuracy
Unit: %
Benchmark version: 2
Category: finance
Task: overall
Task label: Overall
Direction: Higher is better
Comparison cohort: vals-corp-fin-v2:2:overall
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-10-15
Methodology: Source model ID: minimax/MiniMax-M2.1. Source model name: MiniMax M2.1. Provider reported by source: MiniMax. Evaluator: Vals AI. Dataset type: private. Leaderboard date: 2026-08-12. Temperature: 1. Top P: 0.95. Maximum output tokens: 131000. Latency in seconds: 21.969. Standard error: 0.968. Cost per task in USD: 0.022461. Source ID: vals-ai. Evidence capture ID: vals-corp_fin_v2-2026-10-15-7cf2217ab372. Source payload hash: 7cf2217ab3728d5a84bffce82aab9fc52d7f810b0cbe8c65b387df47f3d31c9b. Parser version: vals-astro-v3. Imported at: 2026-10-15
Comparison methodology: Temperature: 1. Top P: 0.95. Maximum output tokens: 131000

Source Vals AI

CorpFin v2: Shared Max Context (Show measurement, test conditions, and source)

Source value: 61.072
Score: 61.072
Metric: accuracy
Unit: %
Benchmark version: 2
Category: finance
Task: shared_max_context
Task label: Shared Max Context
Direction: Higher is better
Comparison cohort: vals-corp-fin-v2:2:shared_max_context
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-10-15
Methodology: Source model ID: minimax/MiniMax-M2.1. Source model name: MiniMax M2.1. Provider reported by source: MiniMax. Evaluator: Vals AI. Dataset type: private. Leaderboard date: 2026-08-12. Temperature: 1. Top P: 0.95. Maximum output tokens: 131000. Latency in seconds: 22.18. Standard error: 1.665. Cost per task in USD: 0.024085. Source ID: vals-ai. Evidence capture ID: vals-corp_fin_v2-2026-10-15-7cf2217ab372. Source payload hash: 7cf2217ab3728d5a84bffce82aab9fc52d7f810b0cbe8c65b387df47f3d31c9b. Parser version: vals-astro-v3. Imported at: 2026-10-15
Comparison methodology: Temperature: 1. Top P: 0.95. Maximum output tokens: 131000

Source Vals AI

GPQA Diamond: Few-Shot CoT (Show measurement, test conditions, and source)

Source value: 75.758
Score: 75.758
Metric: accuracy
Unit: %
Benchmark version: 1
Category: reasoning
Task: diamond_few_shot_cot
Task label: Few-Shot CoT
Direction: Higher is better
Comparison cohort: vals-gpqa:1:diamond_few_shot_cot
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-10-15
Methodology: Source model ID: minimax/MiniMax-M2.1. Source model name: MiniMax M2.1. Provider reported by source: MiniMax. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-09-01. Temperature: 1. Top P: 0.95. Maximum output tokens: 131000. Latency in seconds: 110.423. Standard error: 3.088. Cost per task in USD: 0.009533. Source ID: vals-ai. Evidence capture ID: vals-gpqa-2026-10-15-951eb5224097. Source payload hash: 951eb5224097733a9d09d06c14908ae2980e2e640e539f2c284877c5092d7077. Parser version: vals-astro-v3. Imported at: 2026-10-15
Comparison methodology: Temperature: 1. Top P: 0.95. Maximum output tokens: 131000

Source Vals AI

GPQA Diamond: Zero-Shot CoT (Show measurement, test conditions, and source)

Source value: 81.313
Score: 81.313
Metric: accuracy
Unit: %
Benchmark version: 1
Category: reasoning
Task: diamond_zero_shot_cot
Task label: Zero-Shot CoT
Direction: Higher is better
Comparison cohort: vals-gpqa:1:diamond_zero_shot_cot
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-10-15
Methodology: Source model ID: minimax/MiniMax-M2.1. Source model name: MiniMax M2.1. Provider reported by source: MiniMax. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-09-01. Temperature: 1. Top P: 0.95. Maximum output tokens: 131000. Latency in seconds: 160.401. Standard error: 2.93. Cost per task in USD: 0.013312. Source ID: vals-ai. Evidence capture ID: vals-gpqa-2026-10-15-951eb5224097. Source payload hash: 951eb5224097733a9d09d06c14908ae2980e2e640e539f2c284877c5092d7077. Parser version: vals-astro-v3. Imported at: 2026-10-15
Comparison methodology: Temperature: 1. Top P: 0.95. Maximum output tokens: 131000

Source Vals AI

Recorded measurement details

MedCode: Overall (Show measurement, test conditions, and source)

Source value: 34.083
Score: 34.083
Metric: accuracy
Unit: %
Benchmark version: 1
Category: health
Task: overall
Task label: Overall
Direction: Higher is better
Comparison cohort: vals-medcode:1:overall
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-10-15
Methodology: Source model ID: minimax/MiniMax-M2.1. Source model name: MiniMax M2.1. Provider reported by source: MiniMax. Evaluator: Vals AI. Dataset type: private. Leaderboard date: 2026-10-13. Temperature: 1. Top P: 0.95. Maximum output tokens: 30000. Latency in seconds: 19.548. Standard error: 1.943. Source ID: vals-ai. Evidence capture ID: vals-medcode-2026-10-15-988a6ae1e07e. Source payload hash: 988a6ae1e07e33ce031572ed4a804be11c36ee97a1dd75cbe4740fff3b534f8c. Parser version: vals-astro-v3. Imported at: 2026-10-15
Comparison methodology: Temperature: 1. Top P: 0.95. Maximum output tokens: 30000

Source Vals AI

MedScribe: Overall (Show measurement, test conditions, and source)

Source value: 80.777
Score: 80.777
Metric: accuracy
Unit: %
Benchmark version: 1
Category: health
Task: overall
Task label: Overall
Direction: Higher is better
Comparison cohort: vals-medscribe:1:overall
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-10-15
Methodology: Source model ID: minimax/MiniMax-M2.1. Source model name: MiniMax M2.1. Provider reported by source: MiniMax. Evaluator: Vals AI. Dataset type: private. Leaderboard date: 2026-10-13. Temperature: 1. Top P: 0.95. Maximum output tokens: 30000. Latency in seconds: 53.161. Standard error: 1.831. Cost per task in USD: 0.005087. Source ID: vals-ai. Evidence capture ID: vals-medscribe-2026-10-15-271fc122ddc2. Source payload hash: 271fc122ddc2b0602bd6c7947f65a9a7f0d8cab94f6dc27d62a9380a9317f9c0. Parser version: vals-astro-v3. Imported at: 2026-10-15
Comparison methodology: Temperature: 1. Top P: 0.95. Maximum output tokens: 30000

Source Vals AI

Page 1 of 9

Predecessor and successor

Predecessor: MiniMax M2
Model lineage evidence
MiniMax introduces M2.1 as an update to M2 and documents its improvements over M2 in complex tasks, response efficiency and benchmark results.
MiniMax, "MiniMax M2.1" (2026-09-22)
Successor: MiniMax M2.5
Model lineage evidence
MiniMax's own release post reports M2.5's end-to-end runtime improving to 22.8 minutes from 31.3 minutes, 'a 37% speed improvement compared to M2.1', and frames the sequence as having 'successively released M2, M2.1, and M2.5'.
MiniMax, "MiniMax M2.5: Built for Real-World Productivity" (2026-09-19)

Sources and data date

Type	Evidence and data date
Speed	MiniMax. Output speed after the first streaming chunk, not total response time. Classified as deprecated by Artificial Analysis. Artificial Analysis maintains only the default 10K-input workload there. Retrieved September 8, 2026. Source: Artificial Analysis · Artificial Analysis (retrieved September 8, 2026)Speed measurement methodology
Price	Base API prices without caching or batch discounts. Higher context tiers and other rates are listed under Costs. Source: Pricing · API pricing
Research status	Research date September 8, 2026. 2 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication.
Not found in the inspected sources	knowledge-cutoff Original wording No exact value located in the checked primary model card/API documentation.
Additional source	Model metadata (retrieved September 8, 2026) · MiniMax M2.1 model card · Editorial description reviewed October 2, 2026
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	MiniMax M2.1 release · Editorial description reviewed October 2, 2026
Additional source	OpenRouter: Popularity index (derived), CC BY 4.0
Additional source	Vercel AI Gateway: Popularity index (derived), CC BY 4.0
Additional source	Creative Commons Attribution 4.0
Additional source	MiniMax, "MiniMax M2.5: Built for Real-World Productivity" (2026-09-19)

MiniMax M2.1

Specifications and access

Published benchmarks

Health

MedCode

MedScribe

Measurements without matching peer values

Recorded measurement details

Our tests

SVG short test

MiniMax M2.1

Pelican on a bicycle

Popularity

Published prices

Predecessor and successor

Sources and data date