FitMyLLM
Xiaomi/Mixture of Experts

XMiMo V2.5

MiMo V2.5 — Xiaomi's 310B omnimodal MoE with 15B active: text, image, video and audio in one model. The smaller sibling of MiMo V2.5 Pro, and the one people actually download.

chatcodingreasoningmultilingualvisionaudiomathagentictool_use
310.78B
Parameters (15B active)
1024K
Context length
9
Benchmarks
17
Quantizations
430K
HF downloads
Architecture
MoE
Released
2026-04-22
Layers
48
KV Heads
4
Head Dim
192
Family
mimo

Quantization Options

Context length:
QuantBitsVRAM @ 16KQuality
IQ2_XXS2.38
94.6 GB
92.9 + 1.7 KV
low
IQ2_M2.93
116.0 GB
114.3 + 1.7 KV
low
Q2_K3.16
124.9 GB
123.2 + 1.7 KV
low
IQ3_XXS3.25
128.4 GB
126.7 + 1.7 KV
low
IQ3_XS3.5
138.1 GB
136.5 + 1.7 KV
low
Q3_K_S3.64
143.6 GB
141.9 + 1.7 KV
low
IQ3_M3.76
148.2 GB
146.6 + 1.7 KV
low
Q3_K_M4
157.6 GB
155.9 + 1.7 KV
low
Q3_K_L4.3
169.2 GB
167.5 + 1.7 KV
moderate
IQ4_XS4.46
175.4 GB
173.7 + 1.7 KV
moderate
Q4_K_S4.67
183.6 GB
181.9 + 1.7 KV
moderate
Q4_K_M4.89
192.1 GB
190.5 + 1.7 KV
good
Q5_K_S5.57
218.6 GB
216.9 + 1.7 KV
good
Q5_K_M5.7
223.6 GB
221.9 + 1.7 KV
good
Q6_K6.56
257.0 GB
255.3 + 1.7 KV
excellent
Q8_08.5
332.4 GB
330.7 + 1.7 KV
lossless
FP1616
623.7 GB
622.0 + 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 ~95 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 V2.5 NOW

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

Community Ratings

Loading ratings...

Benchmarks (9)

τ²-Bench90.6
GPQA Diamond84.9
AA Long Context68.3
IFBench67.1
AA Coding56.8
SciCode43.1
Terminal-Bench41.7
AA Intelligence38.0
HLE27.2

Run this model

Easiest way to get started·Beginners
DOCS ↗
curl -fsSL https://ollama.com/install.sh | sh
$ollama run mimo:311b-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 V2.5

Build Hardware for MiMo V2.5
▸ SPEC SHEET

MiMo V2.5310.78B MoE.

▸ SPECIFICATIONS
PARAMETERS
310.78B (15B active)
ARCHITECTURE
Mixture of Experts
CONTEXT LENGTH
1024K tokens
CAPABILITIES
chat, coding, reasoning, multilingual, vision, audio, math, agentic, tool_use
RELEASE DATE
2026-04-22
PROVIDER
Xiaomi
FAMILY
mimo
▸ VRAM REQUIREMENTS
QUANTBPWVRAMQUALITY
IQ2_XXS2.3892.9 GB65%
IQ2_M2.93114.3 GB75%
Q2_K3.16123.2 GB78%
IQ3_XXS3.25126.7 GB82%
IQ3_XS3.5136.5 GB84%
Q3_K_S3.64141.9 GB85%
IQ3_M3.76146.6 GB86%
Q3_K_M4155.9 GB88%
Q3_K_L4.3167.5 GB90%
IQ4_XS4.46173.7 GB92%
Q4_K_S4.67181.9 GB93%
Q4_K_M4.89190.5 GB94%
Q5_K_S5.57216.9 GB96%
Q5_K_M5.7221.9 GB96%
Q6_K6.56255.3 GB97%
Q8_08.5330.7 GB100%
FP1616622.0 GB100%
§ 01BENCHMARK SCORES
GPQA Diamond84.9
HLE27.2
AA Intelligence38.0
AA Coding56.8
aa_ifbench67.1
aa_terminal_bench41.7
aa_tau290.6
aa_scicode43.1
aa_lcr68.3
§ 03COMPATIBLE GPUs
13 @ Q4_K_M