MiniMax M2.5
MiniMax M2.5Rankingin thinking mode?
Alternatives?
CheaperQwen 3.5 Plusunstated+6%-51%$0.12 / $0.69
BetterClaude Sonnet 4.5thinking+5%+1,043%$0.56OpenMiniMax M3unstated+7%same$0.30 / $1.20Memory (4-bit): 137 GB → 256 GB+2CheaperQwen 3.5 Flashunstated+5%-78%$0.065 / $0.26
CheaperGPT-5.6 Lunaunstated+5%-14%$0.20 / $1.20Qwen 3.5 Plusunstated+6%-51%$0.12 / $0.69Qwen 3.5 Flashunstated+5%-78%$0.065 / $0.26GPT-5.6 Lunaunstated+5%-14%$0.20 / $1.20
GLM 5.3 FlashunstatedOpen+5%-55%$0.15 / $0.50Memory (4-bit): 137 GB → 193 GB+5
Mercury 2.5unstated+4%-36%$0.20 / $0.75Claude Sonnet 4.5thinking+5%+1,043%$0.56Claude Fable 5unstated+11%+3,710%$10 / $50Claude Fable 5.1unstated+11%+3,710%$10 / $50Replace forced tool_choice+5
Gemini 4 Argonunstated+10%+662%*$2 / $10Synchronous output stops at 65,536+2Gemini 3.1 Pro Previewunstated+9%+757%$2 / $12Cannot use both thinking controlsMiniMax M3unstatedOpen+7%same$0.30 / $1.20Memory (4-bit): 137 GB → 256 GB+2MiniMax M2.7unstatedOpen+5%same$0.30 / $1.20GLM 5.3 FlashunstatedOpen+5%-55%$0.15 / $0.50Memory (4-bit): 137 GB → 193 GB+5GLM 4.7unstatedOpen+4%+90%$0.60 / $2.20Context: 205K → 203K+1TInkling SmallunstatedOpen+4%+21%$0.45 / $1.20Memory (4-bit): 137 GB → 160 GB+1MiniMax M3unstatedOpen+7%—$0.30 / $1.20Memory (4-bit): 137 GB → 256 GB+2MiniMax M2.7unstatedOpen+5%—$0.30 / $1.20MiniMax M2.1unstatedOpen-6%—$0.30 / $1.20MiniMax M3.1 Flashthinkingn/a—$0.22No strict JSON+1MiniMax M2thinkingOpenn/a—$0.30 / $1.20MiniMax M1 40kOpenn/a——Memory (4-bit): 137 GB → 274 GBMiniMax M1 80kOpenn/a—$0.55 / $2.20Memory (4-bit): 137 GB → 274 GBMiniMax M1unstatedOpenn/a—$0.40 / $2.20Memory (4-bit): 137 GB → 274 GB+1All MiniMax models →Run it yourself
estimateHistory
6 versionsBenchmarks
all categoriesFAQ
How much does MiniMax M2.5 cost?›
$0.30 per 1M input tokens and $1.20 per 1M output tokens; cached input $0.03 per 1M; cheapest via Azure AI Foundry at $0.033 / $1.32.
Is MiniMax M2.5 free to use?›
There is no free API; the cheapest paid price is $0.033 per 1M input tokens. The weights are open, so you can run it yourself.
How good is MiniMax M2.5, and what is it best at?›
#124 of 336 in Chat & Knowledge, #74 of 97 in Coding, #18 of 242 in Agents & Tool Use, #112 of 197 in Writing & Language. Strongest at: Tool calling, Instructions.
Is MiniMax M2.5 better than MiniMax M2.1?›
On Vals LegalBench, MiniMax M2.5 scores 80 vs 75.5 — higher.
How can I use MiniMax M2.5?›
Through the API of AWS Bedrock, AtlasCloud, Azure AI Foundry, DeepInfra, DigitalOcean and 5 more.
What is the context window of MiniMax M2.5?›
205K tokens, with up to 131K output tokens.
When was MiniMax M2.5 released? Is there a newer version?›
Released 2026-02-12. It has been superseded by MiniMax M2.7.
Is MiniMax M2.5 open source?›
Yes, the weights are open (modified-mit licence).
Can I run MiniMax M2.5 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can MiniMax M2.5 read images?›
No, it accepts text only.
What is MiniMax M2.5?›
MiniMax M2.5 is an open-weights language model from MiniMax, released Feb 2026, the successor to MiniMax M2.1.
What are the best alternatives to MiniMax M2.5?›
Qwen 3.5 Plus (cheaper), Claude Sonnet 4.5 (better), MiniMax M3 (open weights).