Skip to content

MiMo V2.6 Flash VRAM requirements

How much VRAM MiMo V2.6 Flash needs
FP4 + FP8 weights: 161 GBWith 8K context: 162 GBWith 32K context: 163 GB
Parameters
309B
Active per token
15B
Longest context
1M
Released
Sep 2026
License
MIT
Strongest at
Coding, Agents

How much VRAM MiMo V2.6 Flash needs

Memory MiMo V2.6 Flash needs at each precision and context length
PrecisionWeightsWith 8KWith 32KWith 128K
As released (FP4 + FP8)161 GB162 GB163 GB165 GB
3-bit141 GB142 GB142 GB144 GB
2-bit108 GB109 GB110 GB112 GB
1.6-bit59.4 GB60.4 GB60.9 GB63.1 GB

One conversation, 16-bit context cache, runtime overhead included.

GPUs and Macs that run MiMo V2.6 Flash

The best way MiMo V2.6 Flash runs on common GPUs, Macs and AI boxes
MachineBest fitContextSpeedFirst replyReport
RTX 4060Too bigPDF: MiMo V2.6 Flash on RTX 4060 report
RTX 5060Too bigPDF: MiMo V2.6 Flash on RTX 5060 report
RTX 3070Too bigPDF: MiMo V2.6 Flash on RTX 3070 report
RTX 3080 10 GBToo bigPDF: MiMo V2.6 Flash on RTX 3080 10 GB report
RTX 3060 12 GBToo bigPDF: MiMo V2.6 Flash on RTX 3060 12 GB report
Arc B580Too bigPDF: MiMo V2.6 Flash on Arc B580 report
RTX 4070Too bigPDF: MiMo V2.6 Flash on RTX 4070 report
RTX 4070 SuperToo bigPDF: MiMo V2.6 Flash on RTX 4070 Super report
RTX 5070Too bigPDF: MiMo V2.6 Flash on RTX 5070 report
RTX 4060 Ti 16 GBToo bigPDF: MiMo V2.6 Flash on RTX 4060 Ti 16 GB report
RTX 5060 Ti 16 GBToo bigPDF: MiMo V2.6 Flash on RTX 5060 Ti 16 GB report
Arc A770 16 GBToo bigPDF: MiMo V2.6 Flash on Arc A770 16 GB report
Radeon RX 9070 XTToo bigPDF: MiMo V2.6 Flash on Radeon RX 9070 XT report
RTX 4070 Ti SuperToo bigPDF: MiMo V2.6 Flash on RTX 4070 Ti Super report
RTX 4080 SuperToo bigPDF: MiMo V2.6 Flash on RTX 4080 Super report
RTX 5070 TiToo bigPDF: MiMo V2.6 Flash on RTX 5070 Ti report
RTX 5080Too bigPDF: MiMo V2.6 Flash on RTX 5080 report
Radeon RX 7900 XTXToo bigPDF: MiMo V2.6 Flash on Radeon RX 7900 XTX report
RTX 3090Too bigPDF: MiMo V2.6 Flash on RTX 3090 report
RTX 4090Too bigPDF: MiMo V2.6 Flash on RTX 4090 report
Radeon AI PRO R9700Too bigPDF: MiMo V2.6 Flash on Radeon AI PRO R9700 report
RTX 5090Too bigPDF: MiMo V2.6 Flash on RTX 5090 report
2× RTX 3090Too bigPDF: MiMo V2.6 Flash on 2× RTX 3090 report
2× RTX 4090Too bigPDF: MiMo V2.6 Flash on 2× RTX 4090 report
RTX PRO 5000 Blackwell 48 GBToo bigPDF: MiMo V2.6 Flash on RTX PRO 5000 Blackwell 48 GB report
RTX 6000 AdaToo bigPDF: MiMo V2.6 Flash on RTX 6000 Ada report
L40SToo bigPDF: MiMo V2.6 Flash on L40S report
2× RTX 5090Too bigPDF: MiMo V2.6 Flash on 2× RTX 5090 report
4× RTX 3090Too bigPDF: MiMo V2.6 Flash on 4× RTX 3090 report
RTX PRO 6000 BlackwellToo bigPDF: MiMo V2.6 Flash on RTX PRO 6000 Blackwell report
Mac mini M4 16 GBToo bigPDF: MiMo V2.6 Flash on Mac mini M4 16 GB report
Mac mini M4 24 GBToo bigPDF: MiMo V2.6 Flash on Mac mini M4 24 GB report
Mac mini M4 32 GBToo bigPDF: MiMo V2.6 Flash on Mac mini M4 32 GB report
Mac mini M4 Pro 24 GBToo bigPDF: MiMo V2.6 Flash on Mac mini M4 Pro 24 GB report
Mac mini M4 Pro 48 GBToo bigPDF: MiMo V2.6 Flash on Mac mini M4 Pro 48 GB report
Mac mini M4 Pro 64 GBToo bigPDF: MiMo V2.6 Flash on Mac mini M4 Pro 64 GB report
M4 Max 128 GBToo bigPDF: MiMo V2.6 Flash on M4 Max 128 GB report
M5 Max 128 GBToo bigPDF: MiMo V2.6 Flash on M5 Max 128 GB report
M3 Ultra 96 GBToo bigPDF: MiMo V2.6 Flash on M3 Ultra 96 GB report
M3 Ultra 256 GBFP4 + FP8928K37 tok/s8.9 sPDF: MiMo V2.6 Flash on M3 Ultra 256 GB report
M3 Ultra 512 GBFP4 + FP81M37 tok/s8.9 sPDF: MiMo V2.6 Flash on M3 Ultra 512 GB report
M5 Ultra 512 GBFP4 + FP81M49 tok/s2.8 sPDF: MiMo V2.6 Flash on M5 Ultra 512 GB report
Ryzen AI Max+ 395 64 GBToo bigPDF: MiMo V2.6 Flash on Ryzen AI Max+ 395 64 GB report
Ryzen AI Max+ 395 128 GBToo bigPDF: MiMo V2.6 Flash on Ryzen AI Max+ 395 128 GB report
RTX Spark2-bit256K34 tok/s3.2 sPDF: MiMo V2.6 Flash on RTX Spark report
DGX Spark2-bit256K31 tok/s3.2 sPDF: MiMo V2.6 Flash on DGX Spark report
Dual DGX SparkFP4 + FP81M17 tok/s2.1 sPDF: MiMo V2.6 Flash on Dual DGX Spark report
DGX Station GB300FP4 + FP81M198 tok/s0.3 sPDF: MiMo V2.6 Flash on DGX Station GB300 report
2× RTX PRO 6000 BlackwellFP4 + FP8896K87 tok/s0.4 sPDF: MiMo V2.6 Flash on 2× RTX PRO 6000 Blackwell report
4× RTX PRO 6000 BlackwellFP4 + FP81M87 tok/s0.2 sPDF: MiMo V2.6 Flash on 4× RTX PRO 6000 Blackwell report
8× RTX PRO 6000 BlackwellFP4 + FP81M87 tok/s0.1 sPDF: MiMo V2.6 Flash on 8× RTX PRO 6000 Blackwell report
A100 80 GB SXMToo bigPDF: MiMo V2.6 Flash on A100 80 GB SXM report
H100 SXMToo bigPDF: MiMo V2.6 Flash on H100 SXM report
4× H200 NVLFP4 + FP81M164 tok/s0.2 sPDF: MiMo V2.6 Flash on 4× H200 NVL report
8× H200 NVLFP4 + FP81M164 tok/s0.1 sPDF: MiMo V2.6 Flash on 8× H200 NVL report

Best fit is the highest quality, up to 8-bit, that leaves 16K of context. First reply is the wait for a 2K prompt. Speeds are rough, plus or minus 25%.

How to run it

llama.cpp
llama-server -hf ggml-org/MiMo-V2.6-Flash-RL-GGUF
vLLM
vllm serve XiaomiMiMo/MiMo-V2.6-Flash-RL
MLX (Mac)
mlx_lm.chat --model mlx-community/MiMo-V2.6-Flash-RL-mxfp4-q8

Questions

How much VRAM does MiMo V2.6 Flash need?

MiMo V2.6 Flash needs 161 GB for its weights as released (FP4 + FP8), 162 GB with an 8K context and 163 GB with 32K.

What is the minimum VRAM for MiMo V2.6 Flash?

About 109 GB at 2-bit with an 8K context, the lowest precision we recommend for a model this size. The smallest machine here that runs it without spilling is the M3 Ultra 256 GB.

How big is MiMo V2.6 Flash?

309B parameters (15B active per token): 161 GB of weights as released (FP4 + FP8).

Can MiMo V2.6 Flash run on an RTX 4090?

Not on one card. The smallest machine here that runs it is the M3 Ultra 256 GB.

More models