MiMo-V2.6-Flash-RL
MiMo-V2.6-Flash-RL is an open-weight language model from Xiaomi MiMo with 311B parameters (mixture of experts). At 4-bit it needs about 206.4 GB of memory, so the smallest device in our directory that fits it is the Rubin, which is announced but not shipping yet.
Overview
Includes room for an 8k context and runtime overhead. More context needs more memory.
Key information
- Lab
- Xiaomi MiMo
- Released
- 21 Sep 2026
- Parameters
- 311B · MoE
- Type
- Language
- Context
- 1,048,576 tokens
- License
- MIT
- HF downloads, 30 days
- 57,506
- Repository
- XiaomiMiMo/MiMo-V2.6-Flash-RL
More from Xiaomi MiMo
View allWill it run on your hardware?
How we estimate| Estimate or device | Memory |
|---|---|
| Q4 | 206.4 GB |
| Q8 | 363.7 GB |
| BF16 | 685 GB |
| RTX 5060 Ti 16 GB | 16 GB usable |
| RTX 5090 32 GB | 32 GB usable |
| DGX Spark 128 GB | 108.8 GB usable |
| Mac Studio M5 Ultra 256 GB | 192 GB usable |
Single-device memory estimate. Pick a device to update the On your device column.
36 GB usable memory
| Format | Memory needed | On your device | Example hardware |
|---|---|---|---|
| 4-bit (Q4) | ≈ 206.4 GB | Too big 573% | RubinAnnounced288 GB memory |
| 8-bit (Q8) | ≈ 363.7 GB | Too big 1010% | No single device in our directory has enough memory |
| 16-bit (BF16) | ≈ 685 GB | Too big 1903% | No single device in our directory has enough memory |
Fits means the estimate stays within 90% of usable memory. Tight is above 90% and up to 100%. Too big is above 100%. These are memory checks, not speed or compatibility tests.
Mac allowances are 66% of total RAM up to 36 GB, and 75% above. AI boxes allow 85% because their memory is shared with the operating system.
Where to run it
On a Mac
Start with the MLX files, built for Apple Silicon.
mlx-community/MiMo-V2.6-Flash-RL-mxfp4-q8Check your MacWith GGUF
Choose a file for llama.cpp or a compatible app.
ggml-org/MiMo-V2.6-Flash-RL-GGUFFind compatible hardwareThrough an API
GMICloud (bf16) is the cheapest provider we found.
$0.112 / $0.224
Input / output per 1M tokens
₹11 / ₹21 at ₹94.29 per $1
Blended: $0.14 (₹13) per 1M. Ranked by a 3:1 mix of input and output tokens.
mlx_lm.generate --model mlx-community/MiMo-V2.6-Flash-RL-mxfp4-q8 \
--prompt "Explain how a transformer works."llama-cli -hf ggml-org/MiMo-V2.6-Flash-RL-GGUF \
-c 8192 -p "Explain how a transformer works."Example commands built from the repository ids above. They need a working install of MLX or llama.cpp. We have not run them; check each repository for available files.
| Precision | |||||
|---|---|---|---|---|---|
| GMICloudCheapest | $0.14₹13 | $0.112₹11 | $0.224₹21 | 1,050,000 | BF16 |
| Darkbloom | $0.145₹14 | $0.10₹9.43 | $0.28₹26 | 1,048,576 | FP4 |
| Cohere | $0.175₹17 | $0.14₹13 | $0.28₹26 | 1,048,576 | FP8 |
| DeepInfra | $0.175₹17 | $0.14₹13 | $0.28₹26 | 1,048,576 | FP8 |
| Xiaomi | $0.175₹17 | $0.14₹13 | $0.28₹26 | 1,048,576 | FP8 |
| Tenstorrent | $0.2062₹19 | $0.165₹16 | $0.33₹31 | 1,048,576 | FP8 |
| Venice | $0.2187₹21 | $0.175₹17 | $0.35₹33 | 1,000,000 | FP8 |
| Makora | $0.225₹21 | $0.18₹17 | $0.36₹34 | 1,048,576 | FP8 |
INR at ₹94.29 per $1. Prices fetched 14 Nov 2026.
More about this model
Frequently asked questions
How much memory does MiMo-V2.6-Flash-RL need? At an 8k context, about 206.4 GB at 4-bit, 363.7 GB at 8-bit and 685 GB at 16-bit (BF16). A longer context needs more.
Can I run MiMo-V2.6-Flash-RL on a Mac? No current Mac fits the 4-bit estimate of 206.4 GB. The smallest Mac in our directory that does is the Mac Studio M3 Ultra (512GB), which is discontinued. An MLX build is available: mlx-community/MiMo-V2.6-Flash-RL-mxfp4-q8.
Does the full context fit? MiMo-V2.6-Flash-RL supports up to 1,048,576 tokens of context. Our memory estimates use 8k, so long prompts need more memory than shown.
What does it cost through an API? The cheapest provider we found is GMICloud, at $0.11 per 1M input tokens and $0.22 per 1M output tokens (₹11 and ₹21 at 94.29 rupees per dollar). Ranked by a 3:1 mix of input and output tokens.
License in plain English
MIT. This license permits commercial use, modification and redistribution, as long as the copyright and license notice are kept. This summary is not legal advice.
OMHQ RigBench
Not yet tested. We show no speed or quality score without a measurement.
Revision history and sources
Last updated . Sources:Hugging Face model card, MLX repository, GGUF repository, OpenRouter.