LOADING DATASHEET
GROUNDED IN REAL HUMAN CONVERSATIONSYOUR BENCH
Mimo V2 Flash
◆ OPEN
by Qiniu
OPEN WEIGHTSLOCALAPIREASONING
RANKED #251 OF 318 ASSISTANTS · OVERALL #383 OF 495 RANKED · VIBE SCORE 5.7 · 45 VOICES ·
COPY BADGE pastes a README snippet with the rank SVG.
MiMo flash model for fast multimodal assistance and agent workflows
HAPPY9.8
8 mentions
SAD5.2
2 mentions
VIBE TRAJECTORY
26 weeks · 28 voices in this window
TOO QUIET
Not enough weekly voices to call a trend yet.
RUN IT ON YOUR OWN MACHINE
Open weights: download Mimo V2 Flash once and it's yours. No subscription, no rate limits, works offline.
309B MoE parameters (15B active), a Q4 quantization is approximately 175 GB to 190 GB
- BARE MINIMUM
Apple Mac Studio M2 Ultra (192 GB Unified Memory)
192 GB UNIFIED MEMORY
~$5,800 USEDToo large for standard consumer GPUs. Runs Q4/Q3 quant in unified RAM at ~10 tok/s with short context
- SWEET SPOT
Workstation with 4x NVIDIA RTX 3090 24GB GPUs plus host offl
192 GB VRAM
~$18,000 NEWQ4 MoE inference across 4x GPUs via vLLM or SGLang, ~28 tok/s, 32k context
- FULL POWER
Multi-GPU Server with 4x NVIDIA H200 141GB (TP4)
564 GB VRAM
~$140,000 NEWNative FP8 / FP16 precision with full 256k context and MTP speculative decoding, ~150 tok/s
LEARN HOW TO SET IT UP
WILL MY BOX RUN MIMO V2 FLASH?
The three cards above are researched picks. Search the machine you actually have — we only say yes if it should reply at a usable speed, not just load the weights and crawl.
HARDWARE PICKS & PRICES ARE RESEARCHED FROM THE LIVE WEB AND REFRESHED AUTOMATICALLY. TREAT THEM AS BALLPARK, NOT GOSPEL.
WHAT CAN YOU ACTUALLY DO WITH IT?
Everyday jobs on each of those rigs: how long Mimo V2 Flash takes, and how good the crowd says it is at that kind of work.
| THE JOB | BARE MINIMUMApple Mac Studio M2 Ultra (192 GB Unified Memory) | SWEET SPOTWorkstation with 4x NVIDIA RTX 3090 24GB GPUs plus host offl | FULL POWERMulti-GPU Server with 4x NVIDIA H200 141GB (TP4) |
|---|---|---|---|
| Write an emaila paragraph or two, drafted from a one-line brief | ~50 S★★★★★ | ~18 S★★★★★ | ~3 S★★★★★ |
| Summarize a documenta long report boiled down to the points that matter | ~2.5 MIN★★★★★ | ~54 S★★★★★ | ~10 S★★★★★ |
| Build a websitea small landing page, markup and styles together | ~13 MIN★★★★★ | ~4.8 MIN★★★★★ | ~53 S★★★★★ |
| Build a backendan API with routes, storage and tests | ~42 MIN★★★★★ | ~15 MIN★★★★★ | ~2.8 MIN★★★★★ |
| Build a gamea playable browser game in one file | ~25 MIN★★★★★ | ~8.9 MIN★★★★★ | ~1.7 MIN★★★★★ |
STARS ARE THE COMMUNITY SCORE FOR THAT KIND OF WORK, DOCKED FOR HOW SQUASHED THE WEIGHTS GET AT EACH TIER. TIMES ARE COMPUTED FROM RESEARCHED SPEEDS. BALLPARK, NOT GOSPEL.
WHAT THE INTERNET SAYS
Not enough discussion about Mimo V2 Flash yet to call it. We found 45 posts but people did not say much either way.
39 POSTS · 6 COMMENTS · REDDIT · HACKER NEWS · LEMMY · BLOG · GITHUB
PEOPLE LOVE
- No fan mail yet. The crowd is still warming up.
PEOPLE COMPLAIN ABOUT MORE
- No complaints yet. The internet is still looking for one. Suspicious.
LOOKING FOR AN ALTERNATIVE? SEE ALL
Other models the crowd has fully reviewed, starting with text models like this one.
Muse Spark 1.3 Contributor
LFM2.5 24B A2B> SOURCES (12)
- REDDIT · 2026-03-23 · 413 UPVOTESA "phone" company is now competing with Anthropic on AI benchmarks. Xiaomi's MiMo-V2-Pro ranks #3 globally on agent tasks.
A "phone" company is now competing with Anthropic on AI benchmarks. Xiaomi's MiMo-V2-Pro ranks #3 globally on agent tasks.
- REDDIT · 2025-12-17 · 106 UPVOTESXiaomi releases "MiMo-V2-Flash" — An Open-Source MoE (309B/15B Active) that hits 150 tokens/s and claims to match DeepSeek-V3.2 & Gemini…
Xiaomi releases "MiMo-V2-Flash" — An Open-Source MoE (309B/15B Active) that hits 150 tokens/s and claims to match DeepSeek-V3.2 & Gemini 3.0 Pro.
- REDDIT · 2025-12-20 · 84 UPVOTESTo further emphasize how busy year this week as been in terms of LLM releases, Xiaomi released their MiMo-V2-Flash open weights language…
To further emphasize how busy year this week as been in terms of LLM releases, Xiaomi released their MiMo-V2-Flash open weights language model, rivaling the likes of DeepSeek 3.2. Its strengths include state-of-an-art agentic tool use.
- REDDIT · 2026-03-23 · 63 UPVOTESXiaomi's MiMo models are making the AI pricing conversation uncomfortable
Xiaomi's MiMo models are making the AI pricing conversation uncomfortable
- REDDIT · 2026-03-01 · 49 UPVOTES[R] Benchmarked 94 LLM endpoints for jan 2026. open source is now within 5 quality points of proprietary
[R] Benchmarked 94 LLM endpoints for jan 2026. open source is now within 5 quality points of proprietary
- BLOG · 2026-08-10 · 32 UPVOTES5 useful things you'll learn in my new post-training textbook (shipping now!)
Housekeeping: No voiceover on another quick “launch” post. More essays soon! After a few long years of finding time to document my lessons from training open models, my post-training book is done! It’s published by Manning, under the title Reinforcement…
- BLOG · 2026-06-16 · 32 UPVOTESFrontier post-training recipe review with Finbarr Timbers
As I’ve been recapping fundamentals of post-training to wrap up my RLHF / Post-training book I knew I needed to get Finbarr Timbers back on the podcast to talk about the state of play. Over the last few months we’ve had many discussions on what we’d need to…
- BLOG · 2026-06-06 · 32 UPVOTESLLM Research Papers: The 2026 List (January to May)
As some of you know, I have the long-running habit of keeping a running list of research papers I want to read, revisit, or cite in future articles and projects. Last year, I shared two organized paper lists, one covering January to June and another one…
- REDDIT · 2026-03-27 · 22 UPVOTESMimo V2 Pro turns out to be very good
Mimo V2 Pro turns out to be very good
- REDDIT · 2025-12-22 · 11 UPVOTESMiMo-V2-Flash. Has anyone tried it yet?
MiMo-V2-Flash. Has anyone tried it yet?
- REDDIT · 2026-01-27 · 6 UPVOTESWhich subscription/api has bang for the buck?
Which subscription/api has bang for the buck?
- REDDIT · 2026-02-24 · 0 UPVOTESWe're heading into the dark ages in terms of free proxies aren't we? (Openrouter)
We're heading into the dark ages in terms of free proxies aren't we? (Openrouter)