MiniMax M2.5
Frontier coding model with 80.2% on SWE-Bench Verified, running at 100 tokens/second at $1/hour—competitive with Claude Opus 4.6 at 10x lower cost.
What is MiniMax M2.5?
MiniMax M2.5 is a frontier LLM designed for real-world productivity, achieving 80.2% on SWE-Bench Verified and 51.3% on Multi-SWE-Bench. Released February 12, 2026, it targets multi-step agent workflows with native tool calling and search capabilities. M2.5 runs at 100 tokens/second (Lightning version) or 50 TPS (standard), with input pricing at $0.3/M tokens and output at $2.4/M—roughly one-tenth to one-twentieth the cost of Claude Opus 4.6, GPT-5, or Gemini 3 Pro. It completes SWE-Bench tasks 37% faster than M2.1, matching Claude Opus 4.6 runtime while costing only 10% per task.
Best for: Cost-sensitive coding tasks · High-throughput agent workflows · Multilingual projects
Developer Stack Fit
Quick read on where MiniMax M2.5 fits in a software team's AI stack. Validate final fit against your codebase, data policy, and deployment model.
- Stack layer
- LLM APIs
- Deployment model
- Open-source deployable
- Open-source status
- Yes or source-available
- API support
- API or integration-friendly
- MCP support
- No MCP signal found
- Security posture
- Review vendor privacy and data retention
- Best use case
- Cost-sensitive coding tasks
Discovery graph
Featured in NeuralStackly paths
Product media
Interface proof
Key Features
- 01
80.2% on SWE-Bench Verified
Competitive with Claude Opus 4.6 on coding
- 02
51.3% on Multi-SWE-Bench
10x cheaper than Claude/GPT/Gemini
- 03
76.3% on BrowseComp
37% faster than M2.1 on SWE-Bench
- 04
100 tokens/second throughput (Lightning)
A core development capability that teams use daily.
- 05
Multilingual coding support (10+ languages)
A core development capability that teams use daily.
- 06
Parallel tool calling
A core development capability that teams use daily.
- 07
Native web search integration
A core development capability that teams use daily.
- 08
Office work capabilities (Word, Excel, PowerPoint)
A core development capability that teams use daily.
Pros & Cons
What stands out
- Excellent coding performance at a fraction of the cost
- Fast inference (100 TPS)
- Supports full development lifecycle (0-to-1, 1-to-10, 10-to-90)
- Multilingual coding strengths
- Open-source option available
Watch outs
- Less brand recognition than OpenAI/Anthropic
- Fewer third-party integrations currently
- Chinese company may affect some users' compliance
Pricing Plans
MiniMax M2.5 Pricing
Choose the perfect plan for your needs. All plans include our core features with different usage limits and advanced capabilities.
Open Source
API Access
Need a Custom Solution?
Looking for enterprise features or custom pricing? Contact MiniMax M2.5 directly for tailored solutions.
Contact SalesMost teams land on the API Access plan.
Alternatives
FAQ
What is MiniMax M2.5 and how does it work?
MiniMax M2.5 is a development tool that frontier coding model with 80.2% on swe-bench verified, running at 100 tokens/second at $1/hour—competitive with claude opus 4.6 at 10x lower cost.. It uses AI to help users improve productivity through analyzing input and generating relevant output.
How much does MiniMax M2.5 cost?
MiniMax M2.5 starts at $0.30/month. They offer a free trial so you can test it before committing.
Does MiniMax M2.5 have a free trial?
Yes — 14-day free trial (credit card required)
What can MiniMax M2.5 do?
Explore by task
More development Tools
Cursor
AI-powered code editor with autonomous agents, multi-model support, and Automations for triggering agents via code changes, Slack, or timers.
TurboQuant
Revolutionary KV cache compression achieving 6x memory reduction and 8x speedup for LLM inference with zero accuracy loss.
Ollama
Local-first LLM runtime for running models on your hardware with local privacy, no per-token API costs, and offline-capable workflows.
Affiliate Disclosure: We may earn a commission when you purchase through links on our site. This doesn't affect our editorial independence or the price you pay.
MiniMax M2.5
From $0.30/month