minimax-m2.5MiniMax
- endpoints
- 10
- regions
- 9
- api
- chat
- released
- February 2026
capabilities 4/8
The same weights cost 1.3x more on aws bedrock than on inceptron ab. That is what the endpoints table is for.
| # | flags | ||||||
|---|---|---|---|---|---|---|---|
| 1 |  Inceptron AB | EU | 197K | $0.28 | $1.10 | $0.03 | zdr |
| 2 |  MiniMax | Global | 200K | $0.30 | $1.20 | $0.06 | zdr |
| 3 |  Nebius AI | Global | 128K | $0.30 | $1.20 | $0.30 | zdr |
| 4 |  AWS Bedrock | us-east-1 | 128K | $0.30 | $1.20 | - | zdr |
| 5 | AWS Bedrock | us-east-2 | 128K | $0.30 | $1.20 | - | zdr |
| 6 | AWS Bedrock | us-west-2 | 128K | $0.30 | $1.20 | - | zdr |
| 7 | AWS Bedrock | eu-central-1 | 128K | $0.36 | $1.44 | - | zdr |
| 8 | AWS Bedrock | eu-north-1 | 128K | $0.36 | $1.44 | - | zdr |
| 9 | AWS Bedrock | eu-south-1 | 128K | $0.36 | $1.44 | - | zdr |
| 10 | AWS Bedrock | eu-west-1 | 128K | $0.36 | $1.44 | - | zdr |
no measured traffic for this model yet. the endpoints table above carries provider pricing
| eval | Score as a share of 100 | /100 |
|---|---|---|
| τ²-Bench | 95.3 | |
| GPQA Diamond | 84.8 | |
| Terminal-Bench Hard | 34.8 | |
| Intelligence Index | 22.8 | |
| Humanity's Last Exam | 20.5 | |
| 5 evals | mean | 51.6 |
Scores from Artificial Analysis and public leaderboards, normalised to 100. Bar colour is the band, not the rank: green 80 and up, blue 55 and up, amber 30 and up. Benchmarks measure narrow skills, so test on your own workload before committing. Released 2026-02-12.
inceptron/minimax-m2.5this model has no managed policy yet, so call the provider endpoint directly. every id in the endpoints table works the same way
Base url is https://router.requesty.ai/v1 for every id here. One key reaches the whole catalog.
from openai import OpenAI client = OpenAI( api_key="YOUR_REQUESTY_API_KEY", base_url="https://router.requesty.ai/v1",) response = client.chat.completions.create( model="inceptron/minimax-m2.5", messages=[ {"role": "user", "content": "Explain quantum computing in one paragraph."}, ],) print(response.choices[0].message.content)
Change the base url, use your Requesty key, set the model to any id in the call pane. Existing OpenAI SDK code needs no other edit, and the same key reaches every other model in the catalog, so switching later is a one-parameter change. Browse all models
MiniMax-M2.5 is a SOTA large language model designed for real-world productivity. Trained in a diverse range of complex real-world digital working environments, M2.5 builds upon the coding expertise of M2.1 to extend into general office work, reaching fluency in generating and operating Word, Excel, and Powerpoint files, context switching between diverse software environments, and working across different agent and human teams. Scoring 80.2% on SWE-Bench Verified, 51.3% on Multi-SWE-Bench, and 76.3% on BrowseComp, M2.5 is also more token efficient than previous generations, having been trained to optimize its actions and output through planning.
questions 5
Which providers serve minimax-m2.5?
How much does minimax-m2.5 cost?
Is minimax-m2.5 open weights?
What is the context window of minimax-m2.5?
How do I use minimax-m2.5 with the OpenAI SDK?
more from minimax 5
| model | context | from /M | providers |
|---|---|---|---|
| minimax-m3 | 1.0M | $0.30 | 3 |
| minimax-m2.7-highspeed | 205K | $0.60 | 2 |
| minimax-m2.7 | 200K | $0.30 | 2 |
| minimax-m2.5-highspeed | 200K | $0.60 | 1 |
| minimax-m2 | 200K | $0.30 | 1 |
route minimax-m2.5 through one endpoint
One key for 4 providers on this model and 600+ others. No markup on provider prices, automatic failover, caching built in. Weekly aggregates in the measured pane come from production traffic routed through Requesty, one line per provider on a shared axis. Methodology