Tsinghua/Zhipu
CogVLM2 19B
当前
6.6K下载量220点赞May 2024发布日期8K tokens上下文Apache 2.0许可证78 优秀质量
CogVLM2 19B (19B parameters) requires approximately 15.5 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 18 GB of VRAM.
快速开始
— 复制粘贴即可本地运行Copy-paste commands to run CogVLM2 19B on your machine.
Run
docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
--hf-repo "THUDM/cogvlm2-llama3-chat-19B" \
--hf-file "cogvlm2-llama3-chat-19B-Q4_K_M.gguf" \
-c 4096 -ngl 99Quick specs
Parameters19B
Architecturedense
Context8K tokens
Modalitytext+vision
Min RAM7.4 GB
Rec. RAM11.6 GB (Q4_K_M)
LicenseApache 2.0
FamilyCogVLM
✓ Vision✓ Chat
About this model
- •Significant improvements in many benchmarks such as TextVQA, DocVQA
- •Support 8K content length
- •Support image resolution up to **1344 * 1344**
- •Provide an open source model version that supports both Chinese and English
快速推荐
最佳硬件
CogVLM2 19B 的最佳选择
运行此模型
量化选项
各量化级别的 VRAM 估算
No hardware detected — fit column shows raw VRAM estimates
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 7.4 GB | Low | — |
Q3_K_S | 3 | 9.3 GB | Low | — |
NVFP4 | 4 | 10.6 GB | Medium | — |
Q4_K_M | 4 | 11.6 GB | Medium | — |
Q5_K_M | 5 | 13.7 GB | High | — |
Q6_K | 6 | 15.6 GB | High | — |
Q8_0 | 8 | 20.3 GB | Very High | — |
F16 | 16 | 38.9 GB | Maximum | — |
硬件兼容性
全部硬件的适配估算
Computing compatibility...
内存详细分析
Reference: RTX 2060 6GB
Weights11.6 GB
KV Cache2.4 GB
Runtime0.9 GB
Headroom0.6 GB
常见问题
FAQ — CogVLM2 19B
另请参阅