ChatGLM-6B
Parameters
6B
Context Length
2K
Modality
Text
Architecture
Dense
KV Cache Rate
448 KB/tok
License
Apache 2.0
Release Date
14 Mar 2023
Knowledge Cutoff
-
API Pricing (per 1M)
Self-hosted only
Hardware Sizing
Precision
Context
Workload Memory: 6.17 GB
INT4 · 2048 context
Base Weights: 2.79 GB
KV Cache: 0.88 GB
Runtime & Graphs: 2.50 GB
Minimum Required Hardware
1x NVIDIA L4 24GB
6.2 / 21.6 GB usable
1x RTX 3060 12GB
6.2 / 10.8 GB usable
Apple M4 24GB
6.2 / 18.0 GB usable
Determine requirements for your specific workload:
Open in VRAM Calculator
Architecture Diagram
Evaluation Benchmarks
Rank
#245
| Benchmark | Score | Rank |
|---|---|---|
General Text | 996 | 188 |
Rankings
Overall Rank
#245
Coding Rank
-
About ChatGLM-6B
ChatGLM-6B is an open-source bilingual dialogue model developed by Tsinghua KEG and Zhipu AI, optimized for consumer-grade GPU deployment. It delivers coherent conversational responses with low memory overhead via INT4 quantization.
Technical Specifications
Attention
Attention Structure
Multi-Head Attention
Attention Heads
32
Key-Value Heads
32
Position Embedding
Absolute Position Embedding
Normalization
Layer Normalization
Activation Function
GELU
Dimensions
Hidden Dimension Size
4,096
Number of Layers
28
FFN Intermediate Size (Dense)
16,384
Tokenizer
Vocabulary Size
130,528
About ChatGLM
ChatGLM series models from Z.ai, based on GLM architecture.