FITMYLLM · AUGUST 17, 2026 · OPEN METHODOLOGY · COMMUNITY BENCHMARKS
LIVE · UPDATED DAILY
309B
Parameters (15B active)
Quantization Options Select your GPU for speed estimates Context length: 4K 8K 16K 32K 64K 128K 256K
Quant Bits VRAM @ 16K Quality IQ2_XXS 2.38 94.1 GB
92.4 + 1.7 KV
low IQ2_M 2.93 115.3 GB
113.7 + 1.7 KV
low Q2_K 3.16 124.2 GB
122.5 + 1.7 KV
low IQ3_XXS 3.25 127.7 GB
126.0 + 1.7 KV
low IQ3_XS 3.5 137.4 GB
135.7 + 1.7 KV
low Q3_K_S 3.64 142.8 GB
141.1 + 1.7 KV
low IQ3_M 3.76 147.4 GB
145.7 + 1.7 KV
low Q3_K_M 4 156.7 GB
155.0 + 1.7 KV
low Q3_K_L 4.3 168.3 GB
166.6 + 1.7 KV
moderate IQ4_XS 4.46 174.4 GB
172.8 + 1.7 KV
moderate Q4_K_S 4.67 182.6 GB
180.9 + 1.7 KV
moderate Q4_K_M 4.89 191.1 GB
189.4 + 1.7 KV
good Q5_K_S 5.57 217.3 GB
215.6 + 1.7 KV
good Q5_K_M 5.7 222.3 GB
220.7 + 1.7 KV
good Q6_K 6.56 255.6 GB
253.9 + 1.7 KV
excellent Q8_0 8.5 330.5 GB
328.8 + 1.7 KV
lossless FP16 16 620.2 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 Chat Coding Reasoning Creative Vision Roleplay Agentic
Loading ratings...
Run this model IQ2_XXS — 92.4 GB VRAM IQ2_M — 113.7 GB VRAM Q2_K — 122.5 GB VRAM IQ3_XXS — 126.0 GB VRAM IQ3_XS — 135.7 GB VRAM Q3_K_S — 141.1 GB VRAM IQ3_M — 145.7 GB VRAM Q3_K_M — 155.0 GB VRAM Q3_K_L — 166.6 GB VRAM IQ4_XS — 172.8 GB VRAM Q4_K_S — 180.9 GB VRAM Q4_K_M — 189.4 GB VRAM Q5_K_S — 215.6 GB VRAM Q5_K_M — 220.7 GB VRAM Q6_K — 253.9 GB VRAM Q8_0 — 328.8 GB VRAM FP16 — 618.5 GB VRAM
Ollama llama.cpp vLLM LM Studio KoboldCpp Jan Docker
▸ Easiest way to get started · Beginners
DOCS ↗ curl -fsSL https://ollama.com/install.sh | shCOPY
$ ollama run mimo:309b-q4_K_MCOPY
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.
Auto-detect GPU Live tok/s in chat Speed benchmarks 9 inference engines
GPUs that can run this model At Q4_K_M quantization. Sorted by minimum VRAM.
Find the best GPU for MiMo V2 Flash 309B
Build Hardware for MiMo V2 Flash 309B ▸ COLOPHON FITMYLLM · INDEPENDENT · DATA-DRIVEN
FITMYLLM · EST. 2025 · © 2026
RECOMMENDATIONS FROM PUBLISHED MATH, CORRECTED BY THE COMMUNITY — 30.
▸ SPEC SHEET
MiMo V2 Flash 309B — 309B 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
QUANT BPW VRAM QUALITY IQ2_XXS 2.38 92.4 GB 65% IQ2_M 2.93 113.7 GB 75% Q2_K 3.16 122.5 GB 78% IQ3_XXS 3.25 126.0 GB 82% IQ3_XS 3.5 135.7 GB 84% Q3_K_S 3.64 141.1 GB 85% IQ3_M 3.76 145.7 GB 86% Q3_K_M 4 155.0 GB 88% Q3_K_L 4.3 166.6 GB 90% IQ4_XS 4.46 172.8 GB 92% Q4_K_S 4.67 180.9 GB 93% Q4_K_M 4.89 189.4 GB 94% Q5_K_S 5.57 215.6 GB 96% Q5_K_M 5.7 220.7 GB 96% Q6_K 6.56 253.9 GB 97% Q8_0 8.5 328.8 GB 100% FP16 16 618.5 GB 100%
§ 01 BENCHMARK SCORES
MMLU-PRO 84.9
LiveCodeBench 80.6
SWE-bench 73.4
AIME 94.1
MATH-500 67.7
GPQA Diamond 83.7
HLE 22.1
§ 03 COMPATIBLE GPUs
13 @ Q4_K_M Feedback