MiMo-V2.6-Flash
Xiaomi · Open weights · Released Sep 21, 2026 · xiaomi/mimo-v2.6-flash
Best $0.10 in / $0.224 out via Darkbloom · fp4 · 1.05M ctx at this price (headline 1.05M)
1.05M contextVisionToolsReasoningStructured outputPrompt cachingAudio inputOpen weights
Available from 7 providersUSD / MTOK · PRICES 23 MINUTES AGO · SPEED 3 DAYS AGO
EST. MONTHLY COSTclick a row for detail · click headers to sort
| PROVIDER | INPUT | OUTPUT | CACHE | CONTEXT | QUANT | TTFT | TPS | UPTIME |
|---|---|---|---|---|---|---|---|---|
 Darkbloom | $0.10 | $0.28 | $0.05 | 1.05M | fp4 | 12.4s | 38 tok/s | 99.8% |
 GMICloud | $0.112 | $0.224 | $0.002 | 1.05M | bf16 | 2.7s | 43 tok/s | 98.7% |
 Xiaomi1ST PARTY | $0.14 | $0.28 | $0.003 | 1.05M | fp8 | 2.7s | 36 tok/s | 100.0% |
 Cohere | $0.14 | $0.28 | $0.003 | 1.05M | — | 0.49s | 125 tok/s | 99.9% |
 DeepInfra | $0.14 | $0.28 | $0.003 | 1.05M | fp8 | ✕ | ✕ | 97.4% |
| MAMakora | $0.14 | $0.28 | $0.003 | 1.05M | fp8 | 0.86s | 191 tok/s | 99.9% |
 Venice | $0.175 | $0.35 | $0.004 | 1M | fp8 | 2.5s | 34 tok/s | 96.6% |
TTFT/TPS: median (P50) over trailing 504h · chat10k profile (~10K in / 200 out) · measured via OpenRouter routing · ✕ = probed, endpoint returned no measurable stream · methodology v1.1 · measured 3 days ago
SERVE THIS MODEL AND NOT LISTED? LIST YOUR INFERENCE →
Input price history · best listed price, observed
↓ 17% SINCE OCT 4, 2026
$0.12 · OCT 4, 2026$0.10 · TODAY
Pricing variants · USD / MTok in / out
STANDARD
$0.10 / $0.28
per MTok in / out
CACHE READ $0.05
Recent changesRSS ↗
Nov 5, 2026endpoint removed · cheaperinference8 → 7
Nov 4, 2026endpoint removed · io-net9 → 8
Nov 3, 2026endpoint removed · novita10 → 9
Nov 2, 2026endpoint removed · relace11 → 10
Oct 30, 2026price decreased · input0.1 → 0.087
Oct 30, 2026price decreased · output0.238 → 0.187
Oct 27, 2026price increased · input0.091 → 0.1
Oct 27, 2026price increased · output0.182 → 0.238
About
MiMo-V2.6-Flash is an open-source foundation model developed by Xiaomi. Built on a Mixture-of-Experts architecture with 309B total parameters and 15B activated per token, it employs a hybrid attention mechanism for...
Specifications
Context window1,050,000
Max output131,072
Modalities intext, image, video, audio
Modalities outtext
LicenseOpen weights
ReleasedSep 21, 2026
TokenizerOther
WeightsHugging Face ↗
Related
Embed badge
[](https://modelindex.ai/models/xiaomi/mimo-v2.6-flash)Sources
OpenRouter API23 minutes ago
AGGREGATORHugging Face23 minutes ago
COMMUNITYModelIndex benchmarks15 days ago
MEASUREDVerified 23 minutes ago · list prices, not negotiated rates.