MiniMax M2.7 highspeed

minimax/minimax-m2.7-highspeed
Flagship
Reasoning
by minimax · 2026-03-18

MiniMax M2.7 high-speed — same model + same 200k context as M2.7, faster output (~100 tps vs ~60 tps).

ctx204.8K tokens
Max output2K
Inputtext
Outputtext
p50 TTFT1.00 s
INPUT$0.60/ 1M tokens
OUTPUT$2.40/ 1M tokens
p50 TTFT1.00 s7d
p95 TTFT2.66 s7d
TRAFFIC31.6Ktokens / 7d

MiniMax M2.7 highspeed is a flagship text-only model developed by MiniMax, a Chinese AI company. It is optimized for fast inference while maintaining strong reasoning capabilities. The model supports…

What is MiniMax M2.7 highspeed?

Who should use this model?

How does it differ from standard MiniMax models?

What are the key specifications?

Code samples

Call from any SDK

OpenAI-compatible — keep the SDK you already use

  • OpenAI SDKhttps://api.orcarouter.ai/v1
from openai import OpenAI

client = OpenAI(
    base_url="https://api.orcarouter.ai/v1",
    api_key="$ORCAROUTER_API_KEY",
)

response = client.chat.completions.create(
    model="minimax/minimax-m2.7-highspeed",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Supported parameters

  • max_completion_tokens
  • reasoning
  • reasoning_split
  • stream
  • temperature
  • top_p

Pricing

Input / 1M tokens$0.600
Output / 1M tokens$2.40
Cache read / 1M$0.060
Cache write / 1M$0.375
CurrencyUSD

Cost calculator

Tokens / month10MM
Input share70%%
Estimated / month $11.40 · With prompt caching $9.51

Estimate based on list price

Token & cost estimator

Input tokens: 20Cost per request: $0.001212

Estimate only — actual token counts depend on the provider's tokenizer.

Performance

p50 TTFT
1.00 s
Output speed
72.2 tok/s
p95 TTFT
2.66 s
Error rate
0%

Public benchmarks

49.6
AA Coding
Better than 69% of models compared
#33 of 106
55.6
AA Intelligence
Better than 75% of models compared
#28 of 110
58.6
AA Math
Better than 38% of models compared
#50 of 81
GPQA Diamond
49.6 index
Humanity's Last Exam
28.1
IFBench
75.7
Long-Context Recall
68.7
MMLU-Pro
62.6 index
SciCode
47.0
TerminalBench Hard
39.4
τ²-Bench
45.6 index
Source: artificialanalysis.ai

How it compares

MiniMax M2.7 highspeedMiniMax M3MiniMax M2.7MiniMax M2.5
Input $/M$0.60$0.30$0.30$0.30
Output $/M$2.40$1.20$1.20$1.20
Context205K1.0M205K205K
Quality8/109/108/107/10
Compare side-by-sideCompare side-by-sideCompare side-by-sideCompare side-by-side

FAQ

How much does MiniMax M2.7 highspeed cost per token?
Input tokens are $0.60 per 1M tokens. Output tokens are $2.40 per 1M tokens. There is no markup; OrcaRouter bills at the provider's rate.
What is the context window size?
The model supports a context window of 204,800 tokens. Max output is 2,048 tokens per request.
What are the model's main strengths?
It achieves an 87.4 score on GPQA Diamond, a graduate-level reasoning benchmark, and offers fast inference with a large context window at a competitive price.
How does it compare to GPT-4 Turbo?
MiniMax M2.7 highspeed is text-only versus GPT-4 Turbo's vision support. It has a larger context (204K vs 128K) and is significantly cheaper ($0.60/$2.40 vs $10/$30 per million tokens). Both are strong on reasoning benchmarks.
Does OrcaRouter handle data privacy?
OrcaRouter acts as a pass-through API. Data is sent to MiniMax's servers for inference. Refer to MiniMax's privacy policy and OrcaRouter's terms for details on data handling and storage.
How do I call this model via an OpenAI-compatible API?
Set your API endpoint to https://api.orcarouter.ai/v1, use the model ID "minimax/minimax-m2.7-highspeed", and pass a standard chat completion request with your OrcaRouter API key.
Is this model suitable for real-time applications?
Yes, the "highspeed" variant is optimized for low latency, making it appropriate for real-time chatbots, live document analysis, and interactive coding assistants.
Can it process images or audio?
No. It is a text-only model. All input must be text. For multimodal tasks, consider models like GPT-4 Vision or Claude 3 Opus.
What is the maximum output length?
The model can generate up to 2,048 tokens per request. For longer outputs, you would need to chunk the generation or use a different model.
Are there any discounts for high volume?
No volume discounts are mentioned. Pricing is linear per token. You pay the same rate regardless of usage volume.

Embed this badge

MiniMax M2.7 highspeed$0.60/M in1000ms p50via OrcaRouter
HTML <a href="https://www.orcarouter.ai/models/minimax/minimax-m2.7-highspeed" target="_blank"> <img src="https://www.orcarouter.ai/embed/minimax/minimax-m2.7-highspeed.svg" alt="MiniMax M2.7 highspeed on OrcaRouter" /> </a>
Markdown [![MiniMax M2.7 highspeed](https://www.orcarouter.ai/embed/minimax/minimax-m2.7-highspeed.svg)](https://www.orcarouter.ai/models/minimax/minimax-m2.7-highspeed)

Model card as data

GET /api/public/models/minimax/minimax-m2.7-highspeedOpen
Machine-readable:/llms.txt/llms-full.txt