MiniMax M2-her
MiniMax M2-herCompare inonEQ-Bench Longform
Rankingconfiguration unstated?
Full ranking →···
82Qwen3 30B A3B35.683MiniMax M2-her34.884Gemma 3 4B34.4Alternatives?
ModelScoreCostPer taskWhat changes
CheaperGPT-5.6 Lunaunstated×2.2-14%$0.20 / $1.20
BetterClaude Opus 5unstated×2.5+1,805%$5 / $25
OpenGLM 5.3unstated×2.4+310%$1.40 / $4.40Reasoning cannot be disabled+3
CheaperDeepSeek V3.1unstatedOpen+79%-19%$0.25 / $0.95CheaperDeepSeek V3 0324unstatedOpen+73%-23%$0.24 / $0.90ModelScoreCostPer taskWhat changes
GPT-5.6 Lunaunstated×2.2-14%$0.20 / $1.20DeepSeek V3.1unstatedOpen+79%-19%$0.25 / $0.95DeepSeek V3 0324unstatedOpen+73%-23%$0.24 / $0.90GPT-5.4 Nanounstated+66%-12%$0.20 / $1.25
Qwen3 235B A22B Instruct 2507unstatedOpen+65%-61%$0.09 / $0.55ModelScoreCostPer taskWhat changes
Claude Opus 5unstated×2.5+1,805%$5 / $25Claude Fable 5.1unstated×2.5+3,710%$10 / $50Replace forced tool_choice+5Claude Opus 5.5unstated×2.4+1,424%$4 / $20No forced tool use+5Claude Fable 5unstated×2.4+3,710%$10 / $50
Grok 4.7unstated×2.4+471%$2 / $6Reasoning cannot be disabled+4ModelScoreCostPer taskWhat changes
GLM 5.3unstatedOpen×2.4+310%$1.40 / $4.40Reasoning cannot be disabled+3
Kimi K3unstatedOpen×2.3+1,043%$3 / $15Sampling parameters are fixed+3Kimi K2.6unstatedOpen×2.3+226%$0.95 / $4GLM 5.2unstatedOpen×2.2+310%$1.40 / $4.40DeepSeek V4 Pro 0813unstatedOpen×2.2+277%$1.32 / $3.96Pass back reasoning_content · Chat Completions+4ModelScoreCostPer taskWhat changes
MiniMax M2.5unstatedOpen+88%—$0.30 / $1.20MiniMax M2.1unstatedOpen+66%—$0.30 / $1.20MiniMax M2unstatedOpen+33%—$0.30 / $1.20MiniMax M3thinkingOpenn/a—$0.51Deployment status experimental · Self-hostedMiniMax M2.7thinkingOpenn/a—$0.30 / $1.20MiniMax M1 40kOpenn/a——MiniMax M1 80kOpenn/a—$0.55 / $2.20MiniMax M1unstatedOpenn/a—$0.40 / $2.20All MiniMax models →Benchmarks
in Writing & LanguageFAQ
How much does MiniMax M2-her cost?›
$0.30 per 1M input tokens and $1.20 per 1M output tokens; cached input $0.03 per 1M.
Is MiniMax M2-her free to use?›
There is no free API; the cheapest paid price is $0.30 per 1M input tokens.
How good is MiniMax M2-her, and what is it best at?›
#83 of 95 in Writing & Language.
Is MiniMax M2-her better than Qwen3 30B A3B?›
On EQ-Bench Longform Creative Writing 1.11, MiniMax M2-her scores 34.8 vs 35.6 — about the same.
How can I use MiniMax M2-her?›
Through the API of Minimax.
What is the context window of MiniMax M2-her?›
66K tokens, with up to 2K output tokens.
Can MiniMax M2-her read images?›
No, it accepts text only.
What is MiniMax M2-her?›
MiniMax M2-her is a language model from MiniMax.
What are the best alternatives to MiniMax M2-her?›
GPT-5.6 Luna (cheaper), Claude Opus 5 (better), GLM 5.3 (open weights).