GLM-5.2, MiniMax-M3 and Kimi K2.7 are now live. Try them now
Sign Up
minimax

minimax-m2.5

Strong general-purpose model from MiniMax, well suited to reasoning and long-context chat.

minimax 64K context
Get API Key View Pricing

Value rank

#7 of 23

intelligence per dollar in our catalogue

Context rank

#8 of 24

64K token window

Agentic strength

95%

tau2-bench tool-use success

Benchmarks

Independent scores by Artificial Analysis, compared with the strongest models in our catalogue.

Intelligence index

glm-5.2
51.1
minimax-m3
44.4
deepseek-v4-pro
44.3
kimi-k2.6
42.8
minimax-m2.5
33.7

Coding index

glm-5.2
50.7
deepseek-v4-pro
47.5
kimi-k2.6
47.1
minimax-m3
43.4
minimax-m2.5
37.4

Agentic (tau2-bench)

deepseek-v4-pro
96.2%
kimi-k2.6
95.9%
minimax-m2.5
95.3%
minimax-m3
88.9%

Best for

Where this model earns its keep.

Agentic pipelines and tool use Prompt-cached workloads

The numbers

Pricing is live from our platform. Prices per 1M tokens, zero data retention on every request.

Input price$0.30
Cache read price$0.08
Output price$1.20
Context window64K tokens
Intelligence / coding index33.7 / 37.4
Agentic: tau2 / terminal-bench95% / 35%
GPQA / MMLU-Pro85% / -

Or consider

Close alternatives in the catalogue.

glm-4.7

Intelligence 33.8 · agentic 96%

glm-5v-turbo

Intelligence 34.5 · agentic 99%

qwen3.5-122b-a10b

Intelligence 32.3 · agentic 94%

Quick start

OpenAI-compatible. Switch in one line.

# pip install openai
client = OpenAI(base_url="https://api.tensorx.ai/v1", api_key="tsx-...")
r = client.chat.completions.create(
    model="minimax/minimax-m2.5",
    messages=[{"role": "user", "content": "Hello"}],
)

Benchmark data from the Artificial Analysis Intelligence Index v4.1, measured independently. Pricing live from the TensorX platform. All inference on EU-sovereign infrastructure with zero data retention.