xiaomi · Open weights
MiMo V2.5 Pro
MiMo V2.5 Pro is Xiaomi's long-context flagship general model for coding and agentic orchestration. It supports tool calling and structured outputs with up to 1M context.
- Reasoning
- Tools
- Structured output
- Input
- Text
- Output
- Text
- Reasoning effort
- none · high
- Discovery tier
- Open weights
Pricing
USD per 1M tokens- Input
- $0.44
- Output
- $0.87
- Cache read
- $0.0036
- Cache write
- $0.00
Reference prices as published by the model provider. API example
Specifications
- Context window
- 1M
- Max output
- 131.1K
- Average output (7 days)
- 596 tokens
- Longer than 55% of 465 models
Side by side
MiMo V2.5 Pro compared
Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.
| MiMo V2.5 ProxiaomiThis model | MiMo V2.5 Pro ThinkingxiaomiRemove | |
|---|---|---|
| Provider | xiaomi | xiaomi |
| Discovery tier | Open weights | Open weights |
| Added | Apr 22, 2026 | Jun 3, 2026 |
| Limits | ||
| Context window | 1.05M | 1.05M |
| Max output | 131.07K | 131.07K |
| Average output | 596 tokens | 1.4K tokens |
| Pricing, USD per 1M tokens | ||
| Input | $0.44 | $0.44 |
| Output | $0.87 | $0.87 |
| Cache read | $0.0036 | $0.0036 |
| Cache write | $0.00 | $0.00 |
| Performance | ||
| Throughput | 21.5 tok/s | 25.9 tok/sBest |
| Time to first token | 4.6 sBest | 5.5 s |
| Request success (24h) | 99.9% | 99.9% |
| Benchmark indexes | ||
| Intelligence Index | 26.0 | 26.0 |
| Coding Index | 60.2 | 60.2 |
| Capabilities | ||
| Reasoning | Yes | Yes |
| Tools | Yes | Yes |
| Structured output | Yes | Yes |
| Reasoning effort | none, high | none, high |
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
- MiMo V2.5 Pro
- MiMo V2.5 Pro Thinking
Intelligence Index
- MiMo V2.5 Pro26.0
- MiMo V2.5 Pro Thinking26.0
MiMo V2.5 Pro: Better than 64% of 326 scored models
Indexes
Coding Index
Composite of the coding evaluations
- MiMo V2.5 Pro: 60.2Top
- MiMo V2.5 Pro Thinking: 60.2Top
Agentic work
AutomationBench-AA
Workflow automation, penalized for guardrail violations
- MiMo V2.5 Pro: 13.7%Top
- MiMo V2.5 Pro Thinking: 13.7%Top
AutomationBench-AA tasks completed
Workflows finished end to end without a guardrail violation
- MiMo V2.5 Pro: 2.0%Top
- MiMo V2.5 Pro Thinking: 2.0%Top
AA-Briefcasev1.1
Agentic knowledge work, rated head to head
- MiMo V2.5 Pro: 883 EloTop
- MiMo V2.5 Pro Thinking: 883 EloTop
GDPval-AAv2.1
Economically valuable occupational tasks, rated head to head
- MiMo V2.5 Pro: 1,121 EloTop
- MiMo V2.5 Pro Thinking: 1,121 EloTop
Document reasoning
GDP.pdf
Professional reasoning over PDF documents, all-pass rate
- MiMo V2.5 Pro: 4.0%Top
- MiMo V2.5 Pro Thinking: 4.0%Top
AA-LCRv1.1
Reasoning across long documents
- MiMo V2.5 Pro: 79.7%Top
- MiMo V2.5 Pro Thinking: 79.7%Top
MLCR-AA
Long-context reasoning over medical records
- MiMo V2.5 Pro: 5.6%Top
- MiMo V2.5 Pro Thinking: 5.6%Top
Reasoning and knowledge
Humanity’s Last Exam
Expert-level questions across academic fields
- MiMo V2.5 Pro: 35.7%Top
- MiMo V2.5 Pro Thinking: 35.7%Top
GPQA Diamond
Graduate-level science questions
- MiMo V2.5 Pro: 86.6%Top
- MiMo V2.5 Pro Thinking: 86.6%Top
IFBench
Following precise output instructions
- MiMo V2.5 Pro: 79.9%Top
- MiMo V2.5 Pro Thinking: 79.9%Top
Coding
Terminal-Benchv4.0
Practical software engineering in a terminal
- MiMo V2.5 Pro: 0.0%Top
- MiMo V2.5 Pro Thinking: 0.0%Top
SciCode
Scientific computing in Python
- MiMo V2.5 Pro: 50.6%Top
- MiMo V2.5 Pro Thinking: 50.6%Top
Legacy evaluations
Terminal-Bench Hard
Earlier terminal task suite
- MiMo V2.5 Pro: 43.2%Top
- MiMo V2.5 Pro Thinking: 43.2%Top
τ²-Bench Telecom
Tool use in a customer-support simulation
- MiMo V2.5 Pro: 94.2%Top
- MiMo V2.5 Pro Thinking: 94.2%Top
AA-LCR (unversioned)
Long-context reasoning, earlier grading
- MiMo V2.5 Pro: 79.7%Top
- MiMo V2.5 Pro Thinking: 79.7%Top
Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis
- MiMo V2.5 Pro
- MiMo V2.5 Pro Thinking
| Leaderboard | MiMo V2.5 Pro | MiMo V2.5 Pro Thinking |
|---|---|---|
| EQ-Bench 4 | #18 · 1201 | Not in top 20 |
Ranks among Noumen catalog models on EQ-Bench leaderboards. Only top-20 placements are listed.EQ-Bench
Live routing data
Performance and reliability
Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.
- Throughput
- 21.5 tok/s
- Faster than 14% of 250 models
- Time to first token
- 4.6 s
- Quicker than 41% of 250 models
- Prompt cache hits
- 78.1%
- About $0.0979 per 1M input tokens with caching
- Request success (24h)
- 99.9%
- Reliable
Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.
- Healthy (98% or more)
- Degraded (90–98%)
- Down (under 90%)
- No requests
Developers
Use it through the API
Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.
curl https://www.noumen.xyz/api/v1/chat/completions \
-H "Authorization: Bearer $NOUMEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "xiaomi/mimo-v2.5-pro",
"reasoning_effort": "none",
"messages": [{ "role": "user", "content": "Explain this stack trace." }]
}'import os
from openai import OpenAI
client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
model="xiaomi/mimo-v2.5-pro",
messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)xiaomi
Related models
Other xiaomi models, starting with the same family.