ChatGLM2-6B
Parameters
6B
Context Length
33K
Modality
Text
Architecture
Dense
KV Cache Rate
28 KB/tok
License
Custom License (ChatGLM2-6B License)
Release Date
25 Jun 2023
Knowledge Cutoff
-
API Pricing (per 1M)
Self-hosted only
Hardware & System Requirements
Precision
Context
Workload Memory: 5.51 GB
INT4 · 8K context
Base Weights: 2.79 GB
KV Cache: 0.22 GB
Runtime & Graphs: 2.50 GB
Minimum Required Hardware
1x NVIDIA L4 24GB
5.5 / 21.6 GB usable
1x RTX 3060 12GB
5.5 / 10.8 GB usable
Apple M4 24GB
5.5 / 18.0 GB usable
Determine requirements for your specific workload:
Open in VRAM Calculator
Runtime requirements
Checkpoint format
FP16
CUDA
>= 11.8
PyTorch
>= 2.1
Transformers
>= 4.27.1
| Architecture | Precision | Attention kernel |
|---|---|---|
Blackwell, Hopper | Native FP16 | FlashAttention-3 |
Ada Lovelace, Ampere | Native FP16 | FlashAttention-2 |
Architecture Diagram
Evaluation Benchmarks
Rank
#248
| Benchmark | Score | Rank |
|---|---|---|
General Text | 1025 | 200 |
Rankings
Overall Rank
#248
Coding Rank
-
About ChatGLM2-6B
ChatGLM2-6B is the second-generation bilingual conversational model by THUDM featuring Multi-Query Attention and FlashAttention. It offers improved inference speed, low latency, and extended context retention on consumer hardware.
Specifications
Attention
Attention Structure
Multi-Head Attention
Attention Heads
32
Key-Value Heads
2
Attention Head Dimension
128
Position Embedding
Absolute Position Embedding
Normalization
RMS Normalization
Activation Function
SwigLU
Dimensions
Base Precision
FP16
Hidden Dimension Size
4,096
Number of Layers
28
FFN Intermediate Size (Dense)
13,696
Tokenizer
Vocabulary Size
65,024
About ChatGLM
ChatGLM series models from Z.ai, based on GLM architecture.