FitMyLLM
OpenBMB/Dense

OMiniCPM5 1B

MiniCPM5 1B — a 1B that scores like a much larger model on instruction following. Small enough to run on a phone.

chatreasoningmultilingual
1.08B
Parameters
128K
Context length
7
Benchmarks
6
Quantizations
1.1M
HF downloads
Architecture
Dense
Released
2026-05-25
Layers
24
KV Heads
2
Head Dim
128
Family
minicpm

Quantization Options

Context length:
QuantBitsVRAM @ 16KQuality
Q4_K_M4.89
1.4 GB
1.1 + 0.3 KV
good
Q5_K_S5.57
1.5 GB
1.2 + 0.3 KV
good
Q5_K_M5.7
1.5 GB
1.3 + 0.3 KV
good
Q6_K6.56
1.7 GB
1.4 + 0.3 KV
excellent
Q8_08.5
1.9 GB
1.6 + 0.3 KV
lossless
FP1616
2.9 GB
2.6 + 0.3 KV
lossless

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

Deploying for a team or in production? Size GPUs, cost & scaling in Enterprise →
READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN MINICPM5 1B NOW

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

Community Ratings

Loading ratings...

Benchmarks (7)

τ²-Bench81.0
IFBench49.3
GPQA Diamond27.8
AA Intelligence11.9
HLE6.5
AA Long Context4.7
SciCode4.4

Run this model

Easiest way to get started·Beginners
DOCS ↗
curl -fsSL https://ollama.com/install.sh | sh
$ollama run minicpm:1b-q4_K_M

Tag may need adjustment — check ollama.com/library/minicpm 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.

NVIDIA Tesla C870
2 GB VRAM • 76.8 GB/s
NVIDIA
NVIDIA Tesla D870
2 GB VRAM • 76.8 GB/s
NVIDIA
NVIDIA Tesla S870
2 GB VRAM • 76.8 GB/s
NVIDIA

Find the best GPU for MiniCPM5 1B

Build Hardware for MiniCPM5 1B
▸ SPEC SHEET

MiniCPM5 1B1.08B Dense.

▸ SPECIFICATIONS
PARAMETERS
1.08B
ARCHITECTURE
Dense Transformer
CONTEXT LENGTH
128K tokens
CAPABILITIES
chat, reasoning, multilingual
RELEASE DATE
2026-05-25
PROVIDER
OpenBMB
FAMILY
minicpm
▸ VRAM REQUIREMENTS
QUANTBPWVRAMQUALITY
Q4_K_M4.891.1 GB94%
Q5_K_S5.571.2 GB96%
Q5_K_M5.71.3 GB96%
Q6_K6.561.4 GB97%
Q8_08.51.6 GB100%
FP16162.6 GB100%
§ 01BENCHMARK SCORES
GPQA Diamond27.8
HLE6.5
AA Intelligence11.9
aa_ifbench49.3
aa_tau281.0
aa_scicode4.4
aa_lcr4.7
§ 03COMPATIBLE GPUs
30 @ Q4_K_M