Qwen3.8 27B is an open-weight dense vision-language model from Qwen.
MiniMax M2.5
MiniMax-M2.5 is a SOTA large language model designed for real-world productivity.
SharkScore
#19 in Models
Why it scores 58.3
MiniMax M2.5 has a SharkScore of 58.3/100, ranked #19 of 608 in Models (updated 2026-11-19).
Each part compares MiniMax M2.5 with other listings in Models. Evidence covers 87% of the score's weight.
- Adoption
- 33 · 30%
- User reviews
- no data · 25%
- Maintenance
- 84 · 20%
- Trust & transparency
- 80 · 15%
- Momentum
- no data · 10%
Paying never changes this score. How ranking works · method v2026-11-18.2
RankShark review of MiniMax M2.5
Automated evaluation by RankShark · 18 Nov 2026 · method v1.0 · How we review
We were not paid for this review.
As of 2026-11-18, MiniMax M2.5 is a text-in/text-out model available through OpenRouter, with a measured 198,000-token context window and usage pricing of $0.295/M input and $1.20/M output. [E1, E2]
MiniMax M2.5 is a text-generation model. The vendor describes it as designed for coding, agentic tool use, search, and office work; OpenRouter lists text input and text output. [E1, E2]
Best for
- Developers evaluating a model for coding tasks. · model card
- Teams building text-based agents that use tools or search. · model card, OpenRouter
- Applications that can use a context window of about 198,000 tokens. · OpenRouter
Not for
- Applications that require image, audio, or other non-text input or output; OpenRouter lists text input and text output. · OpenRouter
- Workflows that require reasoning to be optional; OpenRouter marks reasoning as mandatory. · OpenRouter
Key features
- Coding. The vendor says the model was trained across more than 10 programming languages and over 200,000 real-world environments. · model card
- Agentic tool use and search. The vendor says the model is designed for agentic tool use and search. · model card
- Text-only modalities. OpenRouter lists text as the input and output modality. · OpenRouter
- Context window. RankShark measured a 198,000-token context window on the OpenRouter listing on 2026-11-18. · OpenRouter
- Tool and response controls. OpenRouter lists support for tools, tool choice, structured outputs, and response formats. · OpenRouter
Pros
- The measured context window is 198,000 tokens. · OpenRouter
- The measured OpenRouter rates are $0.295 per million input tokens and $1.20 per million output tokens. · OpenRouter
- The vendor reports coding, tool-use, and search evaluations, including 80.2% on SWE-Bench Verified and 76.3% on BrowseComp with context management. · model card
Cons
- OpenRouter lists text-only input and output, so the listed model modalities do not cover image or audio. · OpenRouter
- OpenRouter marks reasoning as mandatory. · OpenRouter
- No free plan or free trial stated on the pages we checked.
- Whether customer data is used for model training is not stated on the pages we checked.
Pricing
| Plan | Price | Limits |
|---|---|---|
| OpenRouter usage pricing | $0.295 per million input tokens; $1.20 per million output tokens | Pay-per-token rates measured on the OpenRouter listing; no monthly plan limit stated. |
Checked 18 Nov 2026
- Free plan
- Not stated (checked 18 Nov 2026)
- Free trial
- Not stated (checked 18 Nov 2026)
- Trains on customer data
- Not stated (checked 18 Nov 2026)
- Data retention
- Not stated (checked 18 Nov 2026)
- Data regions
- Not stated (checked 18 Nov 2026)
- Data processing agreement
- Not stated (checked 18 Nov 2026)
- AI models used
- MiniMax-M2.5 · model card, OpenRouter
- API
- OpenRouter lists chat completions, completions, messages, and responses endpoints. The vendor page links to a text-generation API. · model card, OpenRouter
- SDKs
- Not stated (checked 18 Nov 2026)
- Webhooks
- Not stated (checked 18 Nov 2026)
- Platforms
- Not stated (checked 18 Nov 2026)
- Security claims
- No security certification stated (checked 18 Nov 2026)
What we measured
- Context window (OpenRouter listing) · now
- 198,000 tokens
- Price per million tokens (OpenRouter listing) · now
- $0.295 in, $1.2 out
- Package downloads · 30 days
- 1,024,915
How we tested · sources
- Hourly uptime and response-time checks of the product's URL.
- Repository, release and download data from GitHub, npm, PyPI and Hugging Face where they exist.
- We did not sign up, pay, or use the product as a customer.
- model card: https://huggingface.co/MiniMaxAI/MiniMax-M2.5 · fetched 18 Nov 2026
- OpenRouter: https://openrouter.ai/minimax/minimax-m2.5 · fetched 18 Nov 2026
- 18 Nov 2026: Plans or prices changed on the pricing page.
- 18 Nov 2026: Plans or prices changed on the pricing page.
Something wrong? Report an error with a public link that shows it; corrections re-run the review.
Top alternatives to MiniMax M2.5
See all- 67.1ModelOpen source
- 65.5
Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per t
ModelOpen source - 64.7
Qwen3.8 Flash is a multimodal reasoning model from Alibaba.
ModelOpen source - 64.3
Qwen3.6 27B is a dense 27-billion-parameter language model from the Qwen Team at Alibaba, released in April 2026.
ModelOpen source