internlm2 20b needs about 14.1 GB of memory at Q4_K_M (a 12.0 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 7 quants, from 8.8 GB (Q3_K_S) to 16.3 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 6.4 GB.
internlm2 20b is a 19.9B-parameter InternLM model in GGUF format, quantized by BoloniniD. Its trained context length is 32K tokens. The repository has 7 quant files, from 8.8 GB (Q3_K_S) to 16.3 GB (Q6_K).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q3_K_S | 8.8 GB | 10.9 GB | Too big | Get file → | |
| Q3_K_M | 9.7 GB | 11.8 GB | Too big | Get file → | |
| Q4_K_S | 11.4 GB | 13.5 GB | Too big | Get file → | |
| Q4_K_M | 12.0 GB | 14.1 GB | Too big | Get file → | |
| Q5_K_S | 13.7 GB | 15.8 GB | Too big | Get file → | |
| Q5_K_M | 14.1 GB | 16.2 GB | Too big | Get file → | |
| Q6_K | 16.3 GB | 18.4 GB | Too big | Get file → |
What GPU do you need for internlm2 20b?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: the best quant that fits is Q3_K_M (9.7 GB file, about 11.8 GB in use).
- 16 GB VRAM: the best quant that fits is Q5_K_S (13.7 GB file, about 15.8 GB in use).
- 24 GB VRAM: the best quant that fits is Q6_K (16.3 GB file, about 18.4 GB in use).
Which internlm2 20b quant should you choose?
Q4_K_M is the usual starting point for internlm2 20b: it balances file size and quality well. Here it is a 12.0 GB file.
If memory is tight, the smallest file is Q3_K_S at 8.8 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about internlm2 20b
How much VRAM does internlm2 20b need?
About 14.1 GB at Q4_K_M with an 8K context window. That is the 12.0 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run internlm2 20b on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 10.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of internlm2 20b GGUF?
It depends on the quant: Q3_K_S is 8.8 GB, Q3_K_M is 9.7 GB, Q4_K_S is 11.4 GB, Q4_K_M is 12.0 GB, Q5_K_S is 13.7 GB, Q5_K_M is 14.1 GB and 1 more.
What context length does internlm2 20b support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.6 GB at 8K and 6.4 GB at 32K.
Where can I download internlm2 20b GGUF?
The files are on Hugging Face in the BoloniniD/internlm2-20b-gguf repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.