FitMyLLM
Zhipu AI/Dense

Zhipu AIGLM-5 753.9B

chat
753.9B
Parameters
198K
Context length
4
Benchmarks
17
Quantizations
136K
HF downloads
Architecture
Dense
Released
2026-02-11
Layers
78
KV Heads
64
Head Dim
64
Family
glm

Quantization Options

QuantBitsVRAMQuality
IQ2_XXS2.38224.8 GBlow
IQ2_M2.93276.6 GBlow
Q2_K3.16298.3 GBlow
IQ3_XXS3.25306.8 GBlow
IQ3_XS3.5330.3 GBlow
Q3_K_S3.64343.5 GBlow
IQ3_M3.76354.8 GBlow
Q3_K_M4377.4 GBlow
Q3_K_L4.3405.7 GBmoderate
IQ4_XS4.46420.8 GBmoderate
Q4_K_S4.67440.6 GBmoderate
Q4_K_M4.89461.3 GBgood
Q5_K_S5.57525.4 GBgood
Q5_K_M5.7537.6 GBgood
Q6_K6.56618.7 GBexcellent
Q8_08.5801.5 GBlossless
FP16161508.3 GBlossless

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

Benchmarks (4)

AIME92.7
GPQA Diamond86.0
SWE-bench77.8
HLE30.5

Run this model

Easiest way to get starteddocs →
curl -fsSL https://ollama.com/install.sh | sh
$ollama run glm:753b-q4_k_m

Downloads and runs automatically. Add --verbose for speed stats.

Setup guide
HuggingFaceOllama LibraryGGUF DownloadsBuild Hardware

Find the best GPU for GLM-5 753.9B

Build Hardware for GLM-5 753.9B

GLM-5 753.9B — 753.9B Parameter Dense LLM

Model Specifications

Parameters
753.9B
Architecture
Dense Transformer
Context Length
198K tokens
Capabilities
chat
Release Date
2026-02-11
Provider
Zhipu AI
Family
glm

VRAM Requirements

QuantizationBPWVRAMQuality
IQ2_XXS2.38224.8 GB65%
IQ2_M2.93276.6 GB75%
Q2_K3.16298.3 GB78%
IQ3_XXS3.25306.8 GB82%
IQ3_XS3.5330.3 GB84%
Q3_K_S3.64343.5 GB85%
IQ3_M3.76354.8 GB86%
Q3_K_M4377.4 GB88%
Q3_K_L4.3405.7 GB90%
IQ4_XS4.46420.8 GB92%
Q4_K_S4.67440.6 GB93%
Q4_K_M4.89461.3 GB94%
Q5_K_S5.57525.4 GB96%
Q5_K_M5.7537.6 GB96%
Q6_K6.56618.7 GB97%
Q8_08.5801.5 GB100%
FP16161508.3 GB100%

Benchmark Scores

GPQA Diamond86.0
SWE-bench77.8
AIME92.7
HLE30.5

How to Run GLM-5 753.9B

Run GLM-5 753.9B locally with Ollama (needs 461.3 GB VRAM at Q4_K_M):

ollama run glm:753b