FitMyLLM
mimo/Mixture of Experts

MMiMo V2 Flash 309B

MiMo-V2-Flash is a Mixture-of-Experts (MoE) language model with 309B total parameters and 15B active parameters.

chatcodingreasoningmathtool_use
309B
Parameters (15B active)
256K
Context length
7
Benchmarks
17
Quantizations
Architecture
MoE
Released
2025-12-16
Layers
48
KV Heads
4
Head Dim
192
Family
mimo

Quantization Options

Context length:
QuantBitsVRAM @ 16KQuality
IQ2_XXS2.38
94.1 GB
92.4 + 1.7 KV
low
IQ2_M2.93
115.3 GB
113.7 + 1.7 KV
low
Q2_K3.16
124.2 GB
122.5 + 1.7 KV
low
IQ3_XXS3.25
127.7 GB
126.0 + 1.7 KV
low
IQ3_XS3.5
137.4 GB
135.7 + 1.7 KV
low
Q3_K_S3.64
142.8 GB
141.1 + 1.7 KV
low
IQ3_M3.76
147.4 GB
145.7 + 1.7 KV
low
Q3_K_M4
156.7 GB
155.0 + 1.7 KV
low
Q3_K_L4.3
168.3 GB
166.6 + 1.7 KV
moderate
IQ4_XS4.46
174.4 GB
172.8 + 1.7 KV
moderate
Q4_K_S4.67
182.6 GB
180.9 + 1.7 KV
moderate
Q4_K_M4.89
191.1 GB
189.4 + 1.7 KV
good
Q5_K_S5.57
217.3 GB
215.6 + 1.7 KV
good
Q5_K_M5.7
222.3 GB
220.7 + 1.7 KV
good
Q6_K6.56
255.6 GB
253.9 + 1.7 KV
excellent
Q8_08.5
330.5 GB
328.8 + 1.7 KV
lossless
FP1616
620.2 GB
618.5 + 1.7 KV
lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

Too big for a single GPU — plan a multi-GPU deployment
Even the lightest quant needs ~94 GB. Size GPUs, replicas, TCO and scaling for a production setup. Open in Enterprise →
READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN MIMO V2 FLASH 309B NOW

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Benchmarks (7)

AIME94.1
MMLU-PRO84.9
GPQA Diamond83.7
LiveCodeBench80.6
SWE-bench73.4
MATH-50067.7
HLE22.1

Run this model

Easiest way to get started·Beginners
DOCS ↗
curl -fsSL https://ollama.com/install.sh | sh
$ollama run mimo:309b-q4_K_M

Tag may need adjustment — check ollama.com/library/mimo for available tags.

▸ SETUP GUIDE
>_

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more
Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines

GPUs that can run this model

At Q4_K_M quantization. Sorted by minimum VRAM.

AMD Instinct MI300X
192 GB VRAM • 5300 GB/s
AMD
$15000
Apple M2 Ultra (192GB)
192 GB VRAM • 800 GB/s
APPLE
$5499
Apple M3 Ultra (192GB)
192 GB VRAM • 800 GB/s
APPLE
$6999
Apple M4 Ultra (192GB)
192 GB VRAM • 1092 GB/s
APPLE
$7499
AMD Radeon Instinct MI300A
192 GB VRAM • 10300 GB/s
AMD
$12000
AMD Radeon Instinct MI300X
192 GB VRAM • 10300 GB/s
AMD
$15000
AMD Radeon Instinct MI308X
192 GB VRAM • 10300 GB/s
AMD
$12000
Apple M5 Ultra (192GB)
192 GB VRAM • 1228 GB/s
APPLE
AMD Radeon Instinct MI325X
288 GB VRAM • 10300 GB/s
AMD
$20000
AMD Radeon Instinct MI350X
288 GB VRAM • 8190 GB/s
AMD
$25000
AMD Radeon Instinct MI355X
288 GB VRAM • 8190 GB/s
AMD
$30000
Apple M4 Ultra (384GB)
384 GB VRAM • 1092 GB/s
APPLE
$9999
Apple M5 Ultra (384GB)
384 GB VRAM • 1228 GB/s
APPLE

Find the best GPU for MiMo V2 Flash 309B

Build Hardware for MiMo V2 Flash 309B
▸ SPEC SHEET

MiMo V2 Flash 309B309B MoE.

▸ SPECIFICATIONS
PARAMETERS
309B (15B active)
ARCHITECTURE
Mixture of Experts
CONTEXT LENGTH
256K tokens
CAPABILITIES
chat, coding, reasoning, math, tool_use
RELEASE DATE
2025-12-16
FAMILY
mimo
▸ VRAM REQUIREMENTS
QUANTBPWVRAMQUALITY
IQ2_XXS2.3892.4 GB65%
IQ2_M2.93113.7 GB75%
Q2_K3.16122.5 GB78%
IQ3_XXS3.25126.0 GB82%
IQ3_XS3.5135.7 GB84%
Q3_K_S3.64141.1 GB85%
IQ3_M3.76145.7 GB86%
Q3_K_M4155.0 GB88%
Q3_K_L4.3166.6 GB90%
IQ4_XS4.46172.8 GB92%
Q4_K_S4.67180.9 GB93%
Q4_K_M4.89189.4 GB94%
Q5_K_S5.57215.6 GB96%
Q5_K_M5.7220.7 GB96%
Q6_K6.56253.9 GB97%
Q8_08.5328.8 GB100%
FP1616618.5 GB100%
§ 01BENCHMARK SCORES
MMLU-PRO84.9
LiveCodeBench80.6
SWE-bench73.4
AIME94.1
MATH-50067.7
GPQA Diamond83.7
HLE22.1
§ 03COMPATIBLE GPUs
13 @ Q4_K_M