tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-30 14:13:57 -04:00
Code Issues Packages Projects Releases Wiki Activity

quantize : update README.md (#14905)

Browse Source 
* Update README.md

* Fix trailing whitespace

* Update README.md

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
This commit is contained in:
Ed Addario
2025-07-27 22:31:11 +01:00
committed by GitHub
parent bf78f5439e
commit 7f97599581
1 changed files with 113 additions and 71 deletions
Download Patch File Download Diff File Expand all files Collapse all files

184
tools/quantize/README.md
View File

@@ -1,18 +1,25 @@
# quantize
This tool takes a GGUF input model file, typically in a high-precision format like F32 or BF16, and converts it to a quantized format.
Quantization reduces the precision of model weights (e.g., from 32-bit floats to 4-bit integers), which shrinks the model's size and can speed up inference.
This process however, may introduce some accuracy loss which is usually measured in [Perplexity](https://huggingface.co/docs/transformers/en/perplexity) (ppl) and/or [Kullback–Leibler Divergence](https://en.wikipedia.org/wiki/Kullback%E2%80%93Leibler_divergence) (kld).
This can be minimized by using a suitable imatrix file.
You can also use the [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space on Hugging Face to build your own quants without any setup.
Note: It is synced from llama.cpp `main` every 6 hours.
Example usage:
```./llama-quantize [options] input-model-f32.gguf [output-model-quant.gguf] type [threads]```
```bash
# obtain the official LLaMA model weights and place them in ./models
# from Hugginface, obtain the official meta-llama/Llama-3.1-8B model weights and place them in ./models
ls ./models
llama-2-7b tokenizer_checklist.chk tokenizer.model
# [Optional] for models using BPE tokenizers
ls ./models
<folder containing weights and tokenizer json> vocab.json
config.json model-00001-of-00004.safetensors model-00004-of-00004.safetensors README.md tokenizer.json
generation_config.json model-00002-of-00004.safetensors model.safetensors.index.json special_tokens_map.json USE_POLICY.md
LICENSE model-00003-of-00004.safetensors original tokenizer_config.json
# [Optional] for PyTorch .bin models like Mistral-7B
ls ./models
<folder containing weights and tokenizer json>
@@ -21,7 +28,7 @@ ls ./models
python3 -m pip install -r requirements.txt
# convert the model to ggml FP16 format
python3 convert_hf_to_gguf.py models/mymodel/
python3 convert_hf_to_gguf.py ./models/mymodel/
# quantize the model to 4-bits (using Q4_K_M method)
./llama-quantize ./models/mymodel/ggml-model-f16.gguf ./models/mymodel/ggml-model-Q4_K_M.gguf Q4_K_M
@@ -37,40 +44,117 @@ Run the quantized model:
./llama-cli -m ./models/mymodel/ggml-model-Q4_K_M.gguf -cnv -p "You are a helpful assistant"
```
When running the larger models, make sure you have enough disk space to store all the intermediate files.
Options:
* `--allow-requantize` allows requantizing tensors that have already been quantized. Warning: This can severely reduce quality compared to quantizing from 16bit or 32bit
* `--leave-output-tensor` will leave output.weight un(re)quantized. Increases model size but may also increase quality, especially when requantizing
* `--pure` disables k-quant mixtures and quantizes all tensors to the same type
* `--imatrix` uses data in file generated by `llama-imatrix` as importance matrix for quant optimizations (highly recommended)
* `--include-weights` use an importance matrix for tensor(s) in the list. Cannot be used with `--exclude-weights`
* `--exclude-weights` use an importance matrix for tensor(s) in the list. Cannot be used with `--include-weights`
* `--output-tensor-type` use a specific quant type for the output.weight tensor
* `--token-embedding-type` use a specific quant type for the token embeddings tensor
* `--keep-split` will generate the quantized model in the same shards as the input file otherwise it will produce a single quantized file
Advanced options:
* `--tensor-type` quantize specific tensor(s) to specific quant types. Supports regex syntax. May be specified multiple times.
* `--prune-layers` prune (remove) the layers in the list
* `--override-kv` option to override model metadata by key in the quantized model. May be specified multiple times
Examples:
```bash
# naive Q4_K_M quantization using default settings and 8 CPU threads. Output will be "ggml-model-Q4_K_M.gguf"
./llama-quantize input-model-f32.gguf q4_k_m 8
```
```bash
# quantize model enabling re-quantization, leaving the output tensor unquantized and all others quantized at the same level (Q4_K)
./llama-quantize --allow-requantize --leave-output-tensor --pure input-model-f32.gguf q4_k_m 8
```
```bash
# quantize model using an importance matrix for specified tensors only (attn_v and ffn_down)
./llama-quantize --imatrix imatrix.gguf --include-weights attn_v --include-weights ffn_down input-model-f32.gguf q4_k_m 8
```
```bash
# quantize model setting output tensor to Q5_K_M, token embeddings to Q3_K_M, and keeping the input file's shards
./llama-quantize --imatrix imatrix.gguf --output-tensor-type q5_k --token-embedding-type q3_k --keep-split input-model-f32.gguf q4_k_m 8
```
```bash
# quantize model using a regex to quantize attn_k tensors in odd layers to Q5_K_M and attn_q tensors in even layers to Q3_K_M
./llama-quantize --imatrix imatrix.gguf --tensor-type "\.(\d*[13579])\.attn_k=q5_k" --tensor-type "\.(\d*[02468])\.attn_q=q3_k" input-model-f32.gguf q4_k_m 8
```
```bash
# quantize model setting tensors attn_v and ffn_down to Q5_K_M and pruning layers 20, 21, and 22
./llama-quantize --imatrix imatrix.gguf --tensor-type attn_v=q5_k --tensor-type ffn_down=q5_k --prune-layers 20,21,22 input-model-f32.gguf q4_k_m 8
```
```bash
# override expert used count metadata to 16, prune layers 20, 21, and 22 without quantizing the model (copy tensors) and use specified name for the output file
./llama-quantize --imatrix imatrix.gguf --override-kv qwen3moe.expert_used_count=int:16 --prune-layers 20,21,22 input-model-f32.gguf pruned-model-f32.gguf copy 8
```
## Memory/Disk Requirements
As the models are currently fully loaded into memory, you will need adequate disk space to save them and sufficient RAM to load them. At the moment, memory and disk requirements are the same.
When running the larger models, make sure you have enough disk space to store all the intermediate files.
As the models are currently fully loaded into memory, you will need adequate disk space to save them and sufficient RAM to load them. At the moment, memory and disk requirements are the same. For exmaple (Llama 3.1):
| Model | Original size | Quantized size (Q4_K_M) |
| ----: | ------------: | ----------------------: |
| 8B | 32.1 GB | 4.9 GB |
| 70B | 280.9 GB | 43.1 GB |
| 405B | 1,625.1 GB | 249.1 GB |
| Model | Original size | Quantized size (Q4_0) |
|------:|--------------:|----------------------:|
| 7B | 13 GB | 3.9 GB |
| 13B | 24 GB | 7.8 GB |
| 30B | 60 GB | 19.5 GB |
| 65B | 120 GB | 38.5 GB |
## Quantization
Several quantization methods are supported. They differ in the resulting model disk size and inference speed.
Several quantization methods are supported. They differ in the resulting model disk size and inference speed. For example,
*(outdated)*
### [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B)
| Model | Measure | F16 | Q4_0 | Q4_1 | Q5_0 | Q5_1 | Q8_0 |
|------:|--------------|-------:|-------:|-------:|-------:|-------:|-------:|
| 7B | perplexity | 5.9066 | 6.1565 | 6.0912 | 5.9862 | 5.9481 | 5.9070 |
| 7B | file size | 13.0G | 3.5G | 3.9G | 4.3G | 4.7G | 6.7G |
| 7B | ms/tok @ 4th | 127 | 55 | 54 | 76 | 83 | 72 |
| 7B | ms/tok @ 8th | 122 | 43 | 45 | 52 | 56 | 67 |
| 7B | bits/weight | 16.0 | 4.5 | 5.0 | 5.5 | 6.0 | 8.5 |
| 13B | perplexity | 5.2543 | 5.3860 | 5.3608 | 5.2856 | 5.2706 | 5.2548 |
| 13B | file size | 25.0G | 6.8G | 7.6G | 8.3G | 9.1G | 13G |
| 13B | ms/tok @ 4th | - | 103 | 105 | 148 | 160 | 131 |
| 13B | ms/tok @ 8th | - | 73 | 82 | 98 | 105 | 128 |
| 13B | bits/weight | 16.0 | 4.5 | 5.0 | 5.5 | 6.0 | 8.5 |
| Measure | IQ1_S | IQ1_M | IQ2_XXS | IQ2_XS | IQ2_S | IQ2_M |
| --------------------------- | ------------ | ------------ | ------------ | ------------- | ------------- | ------------ |
| bits/weight | 2.0042 | 2.1460 | 2.3824 | 2.5882 | 2.7403 | 2.9294 |
| size (GiB) | 1.87 | 2.01 | 2.23 | 2.42 | 2.56 | 2.74 |
| prompt processing t/s @ 512 | 858.88 ±1.22 | 847.99 ±0.47 | 852.39 ±0.85 | 826.99 ±12.51 | 783.55 ±13.73 | 787.68 ±7.00 |
| text generation t/s @ 128 | 79.73 ±0.79 | 72.92 ±0.14 | 79.86 ±0.22 | 78.04 ±0.46 | 77.30 ±2.47 | 74.44 ±0.15 |
| Measure | IQ3_XXS | IQ3_XS | IQ3_S | IQ3_M | IQ4_XS | IQ4_NL |
| --------------------------- | ------------ | ------------ | ------------ | ------------- | ------------- | ------------ |
| bits/weight | 3.2548 | 3.4977 | 3.6606 | 3.7628 | 4.4597 | 4.6818 |
| size (GiB) | 3.04 | 3.27 | 3.42 | 3.52 | 4.17 | 4.38 |
| prompt processing t/s @ 512 | 813.88 ±6.53 | 708.71 ±1.26 | 798.78 ±8.81 | 768.70 ±13.73 | 771.80 ±11.38 | 806.03 ±7.07 |
| text generation t/s @ 128 | 73.95 ±0.20 | 71.67 ±0.54 | 69.31 ±0.63 | 70.15 ±0.33 | 77.51 ±0.20 | 76.63 ±0.28 |
| Measure | Q2_K_S | Q2_K | Q3_K_S | Q3_K_M | Q3_K_L | Q4_K_S |
| --------------------------- | ------------ | ------------ | ------------ | ------------ | ------------ | ------------ |
| bits/weight | 2.9697 | 3.1593 | 3.6429 | 3.9960 | 4.2979 | 4.6672 |
| size (GiB) | 2.78 | 2.95 | 3.41 | 3.74 | 4.02 | 4.36 |
| prompt processing t/s @ 512 | 798.91 ±6.40 | 784.45 ±7.85 | 752.17 ±7.94 | 783.44 ±9.92 | 761.17 ±7.55 | 818.55 ±9.58 |
| text generation t/s @ 128 | 90.01 ±0.12 | 79.85 ±0.20 | 69.84 ±0.18 | 71.68 ±0.22 | 69.38 ±0.49 | 76.71 ±0.20 |
| Measure | Q4_K_S | Q4_K_M | Q5_K_S | Q5_K_M | Q6_K | Q8_0 |
| --------------------------- | ------------ | ------------- | ------------ | ------------ | ------------- | ------------ |
| bits/weight | 4.6672 | 4.8944 | 5.5704 | 5.7036 | 6.5633 | 8.5008 |
| size (GiB) | 4.36 | 4.58 | 5.21 | 5.33 | 6.14 | 7.95 |
| prompt processing t/s @ 512 | 818.55 ±9.58 | 821.81 ±21.44 | 752.52 ±0.99 | 758.69 ±7.43 | 812.01 ±10.82 | 865.09 ±8.30 |
| text generation t/s @ 128 | 76.71 ±0.20 | 71.93 ±1.52 | 69.53 ±0.18 | 67.23 ±1.08 | 58.67 ±3.13 | 50.93 ±0.08 |
| Measure | F16 |
| --------------------------- | ------------ |
| bits/weight | 16.0005 |
| size (GiB) | 14.96 |
| prompt processing t/s @ 512 | 923.49 ±0.53 |
| text generation t/s @ 128 | 29.17 ±0.04 |
## Background information on llama-quantize
- [k-quants](https://github.com/ggml-org/llama.cpp/pull/1684)
- recent k-quants improvements and new i-quants
- k-quants improvements and i-quants
- [#2707](https://github.com/ggml-org/llama.cpp/pull/2707)
- [#2807](https://github.com/ggml-org/llama.cpp/pull/2807)
- [#4773 - 2-bit i-quants (inference)](https://github.com/ggml-org/llama.cpp/pull/4773)
@@ -85,45 +169,3 @@ Several quantization methods are supported. They differ in the resulting model d
- [#5060 - Q3_K_XS](https://github.com/ggml-org/llama.cpp/pull/5060)
- [#5196 - 3-bit i-quants](https://github.com/ggml-org/llama.cpp/pull/5196)
- [quantization tuning](https://github.com/ggml-org/llama.cpp/pull/5320), [another one](https://github.com/ggml-org/llama.cpp/pull/5334), and [another one](https://github.com/ggml-org/llama.cpp/pull/5361)
**Llama 2 7B**
| Quantization | Bits per Weight (BPW) |
|--------------|-----------------------|
| Q2_K | 3.35 |
| Q3_K_S | 3.50 |
| Q3_K_M | 3.91 |
| Q3_K_L | 4.27 |
| Q4_K_S | 4.58 |
| Q4_K_M | 4.84 |
| Q5_K_S | 5.52 |
| Q5_K_M | 5.68 |
| Q6_K | 6.56 |
**Llama 2 13B**
Quantization | Bits per Weight (BPW)
-- | --
Q2_K | 3.34
Q3_K_S | 3.48
Q3_K_M | 3.89
Q3_K_L | 4.26
Q4_K_S | 4.56
Q4_K_M | 4.83
Q5_K_S | 5.51
Q5_K_M | 5.67
Q6_K | 6.56
**Llama 2 70B**
Quantization | Bits per Weight (BPW)
-- | --
Q2_K | 3.40
Q3_K_S | 3.47
Q3_K_M | 3.85
Q3_K_L | 4.19
Q4_K_S | 4.53
Q4_K_M | 4.80
Q5_K_S | 5.50
Q5_K_M | 5.65
Q6_K | 6.56