MiMo-V2.6-Flash
A distilled, low-latency version of MiMo-V2.6 for high-throughput and interactive workloads, still under MIT.
At a glance
Released
May 14, 2026
Provider
Xiaomi
Context window
256,000 tokens
Max output
16,000 tokens
Modalities
Text → text
Knowledge cutoff
January 2026
API model ID
mimo-v2.6-flashLicense
Open weights (MIT)
Pricing
Standard
Input$0.14 / 1M
Output$0.28 / 1M
* Rates are from a third-party aggregator — verify before relying on them. A typical 1,500-token prompt with a 650-token answer costs ≈ $0.0004 on the standard tier. Run your own prompt through the calculator →
What's new vs the previous model
Compared withMiMo-V2.1-Flash (January 2026)
- Faster responses and better short-task accuracy
- Cheaper to serve than the prior Flash
- Drops image input to stay lean
Best use cases
Interactive chat
Powers responsive assistants where latency is the priority.
Bulk processing
Runs classification and extraction over large text volumes cheaply.
Local deployment
Hosts easily under MIT for private, cost-sensitive apps.
Strengths & limitations
Strengths
- Fast
- MIT-licensed
- Inexpensive
Limitations
- Text only
- Less capable than the Pro variant
How to test it
- Download the weights: grab MiMo-V2.6-Flash from Hugging Face (XiaomiMiMo) ↗.
- Run it locally: serve the open weights with vLLM, SGLang or Ollama on your own GPU.
- Call it: hit the local OpenAI-compatible endpoint it exposes (example below).
# serve the open weights, then call the local OpenAI-compatible endpoint
vllm serve XiaomiMiMo/mimo-v2.6-flash
curl http://localhost:8000/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{"model": "mimo-v2.6-flash", "messages": [{"role": "user", "content": "Explain tokenization in one paragraph."}]}'