Model
Model explorer

ChatGLM-6B

OPEN
Zhipu AI / THUDM (Tsinghua) · ChatGLM family · released Mar 14, 2023

Pivotal open bilingual (EN/ZH) dialogue model runnable on consumer GPUs via INT4 (6GB); China's early open ChatGPT-style release.

ReasoningCodingVisionFunction callingTool useAgentic
-456.9
Elo · rank #446
Parameters
6.2B
Active params
6.2B (dense)
Context
2K tokens
Architecture
Dense Transformer (GLM, prefix-LM decoder)
License
Apache-2.0 (code); ChatGLM-6B Model License for weights (free for research; free commercial use after registration)
Languages
2+
API price (in/out)
No hosted API
Modalities
text
Benchmark results
Bar shows position within the tracked field; marker = field best
BIG-Bench HardReasoning18.7%#139
best: ERNIE 4.5 300B-A47B · 94.3%
C-EvalKnowledge38.9%#53
best: Qwen3.6-Plus · 93.3%
CMMLUKnowledge25.3%#32
best: Doubao-1.5-Pro · 90.9%
DROPReasoning3.9#77
best: Hunyuan-T1 · 93.1
GSM8KMath4.8%#185
best: Llama 3.1 405B · 96.8%
HellaSwagReasoning30.4%#154
best: Claude 3 Opus · 95.4%
HumanEvalCoding0.0%#187
best: Claude Opus 4.5 · 99.4%
MATH-500Math3.1%#201
best: GPT-5 · 99.4%
MMLUKnowledge40.6%#257
best: OpenAI o3 · 92.9%
PIQAReasoning65.7%#78
best: GPT-4o mini · 93.1%
Run it locally
VRAM @ Q4
6 GB
VRAM @ FP16
13 GB
Fits on (Q4)
RTX 3060 12GBRTX 4070 Ti 16GBRTX 3090 24GBRTX 4090 24GBRTX 5090 32GBM4 Pro 48GBM3 Max 128GBM3 Ultra 512GBA100 80GBH100 80GBH200 141GBB200 192GB
Throughput data unavailable.
Quantizations
INT4 · INT8
ChatGLM family
Elo progression across releases
ChatGLM-6BMar 2023-456.9ChatGLM2-6BJun 2023-219.2ChatGLM3-6BOct 2023367.9
Resources
HFTHUDM/chatglm-6bGHzai-org/ChatGLM-6B
API price weights · each benchmark row carries its own source badge (see methodology)