bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-07-25 02:31:07 +08:00
commit 6877c431d0
28 changed files with 321 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

47
.gitattributes vendored Normal file
View File

@@ -0,0 +1,47 @@
*.7z filter=lfs diff=lfs merge=lfs -text
*.arrow filter=lfs diff=lfs merge=lfs -text
*.bin filter=lfs diff=lfs merge=lfs -text
*.bin.* filter=lfs diff=lfs merge=lfs -text
*.bz2 filter=lfs diff=lfs merge=lfs -text
*.ftz filter=lfs diff=lfs merge=lfs -text
*.gz filter=lfs diff=lfs merge=lfs -text
*.h5 filter=lfs diff=lfs merge=lfs -text
*.joblib filter=lfs diff=lfs merge=lfs -text
*.lfs.* filter=lfs diff=lfs merge=lfs -text
*.model filter=lfs diff=lfs merge=lfs -text
*.msgpack filter=lfs diff=lfs merge=lfs -text
*.onnx filter=lfs diff=lfs merge=lfs -text
*.ot filter=lfs diff=lfs merge=lfs -text
*.parquet filter=lfs diff=lfs merge=lfs -text
*.pb filter=lfs diff=lfs merge=lfs -text
*.pt filter=lfs diff=lfs merge=lfs -text
*.pth filter=lfs diff=lfs merge=lfs -text
*.rar filter=lfs diff=lfs merge=lfs -text
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
*.tar.* filter=lfs diff=lfs merge=lfs -text
*.tflite filter=lfs diff=lfs merge=lfs -text
*.tgz filter=lfs diff=lfs merge=lfs -text
*.xz filter=lfs diff=lfs merge=lfs -text
*.zip filter=lfs diff=lfs merge=lfs -text
*.zstandard filter=lfs diff=lfs merge=lfs -text
*.tfevents* filter=lfs diff=lfs merge=lfs -text
*.db* filter=lfs diff=lfs merge=lfs -text
*.ark* filter=lfs diff=lfs merge=lfs -text
**/*ckpt*data* filter=lfs diff=lfs merge=lfs -text
**/*ckpt*.meta filter=lfs diff=lfs merge=lfs -text
**/*ckpt*.index filter=lfs diff=lfs merge=lfs -text
*.safetensors filter=lfs diff=lfs merge=lfs -text
*.ckpt filter=lfs diff=lfs merge=lfs -text
*.gguf* filter=lfs diff=lfs merge=lfs -text
*.ggml filter=lfs diff=lfs merge=lfs -text
*.llamafile* filter=lfs diff=lfs merge=lfs -text
*.pt2 filter=lfs diff=lfs merge=lfs -text
*.mlmodel filter=lfs diff=lfs merge=lfs -text
*.npy filter=lfs diff=lfs merge=lfs -text
*.npz filter=lfs diff=lfs merge=lfs -text
*.pickle filter=lfs diff=lfs merge=lfs -text
*.pkl filter=lfs diff=lfs merge=lfs -text
*.tar filter=lfs diff=lfs merge=lfs -text
*.wasm filter=lfs diff=lfs merge=lfs -text
*.zst filter=lfs diff=lfs merge=lfs -text
*tfevents* filter=lfs diff=lfs merge=lfs -text

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-IQ2_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3ab4adbfaa0a7425beec969568c6cff8b6e6cb4b0cd011117b4dae3c807687b8
size 1229034240

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-IQ3_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b428e3a451062ddf6a55f047e69179bcce11b625595a22f5c4c69142e1b4f930
size 1599671040

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-IQ3_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:9320e9a393a2cf3baee62a5ad363d3a6630e73588cb9d31986f4780c77c2733e
size 1476791040

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-IQ3_XXS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b594d5175c2bfdcc0d39ad7a0feea1960cd0beb4ecb0b83a8b982dbdd09873d3
size 1348768512

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-IQ4_NL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:4b4464a1556909f123136504a59134c73f39bbed18c45b6d90dfc71683560e00
size 1917192960

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-IQ4_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:663e14c5e4c45de9c6d21b71139c55a189e0e62234aa31669c669bd6871193b4
size 1829112576

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q2_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:43d601cef7044f6763550243ac2af4dc4dffb3cb7e7742195f1395468b959441
size 1363938048

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q2_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:e73f23f66a467adcc04042ed9fca40423234986b533c82f7125babef03654e49
size 1459360512

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q3_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:6190336b2e744fea83d8c1d5a0d2e2779c89ae88bcc3fc6f00c1033edf61a800
size 1815350016

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q3_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:f25201591b201e0f871a110c1404421b243661e37bcd4c589b40676f9c57c0b8
size 1687161600

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q3_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:501f5f50a6523f73416b09f8e8b038fc903f98a613dc1fb298069680844c3f73
size 1542851328

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q3_K_XL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:a60b0d5dc12b7f2d93c1c814a7b2b332881b63e1a1b65e7976a073a057138b35
size 1910772480

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q4_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:4b2473f0441778aef773fe421c0aa58fef44c9e14ae4423904df0540572a9183
size 1921911552

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q4_1.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:faaa5be865565edb5e001a4d765465f2a4c132b60e6c0618e1cafda15eb51e3a
size 2093353728

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q4_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:e323ff74e852e1c10426d3ff46482fdf4e3ec2698e01aefe80291fbdbeb478ec
size 2114802432

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q4_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:73d9a588383946dcac545a097c47d634558afd79ea43aac3a4563c311d89f195
size 2019379968

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q4_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c73170a950190ed11e2648cec7523b0bedb2da6f723d0eef31d94225b1a97d59
size 1928203008

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q5_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ea0d694905ea5a4e3984064ff12b39f8da461af6724275450ef692758031ab91
size 2417578752

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q5_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c13ca38a96de690759f45ad00da4b7b1d81ca79b26bd3c80976c3d48246caf83
size 2322156288

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q5_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:226ce9c81253c19b587755a5c823a34433a96f4fcedab073852ef1d21c7e792a
size 2269514496

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q6_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:65cac31591244c232d00ef3ff85437fbc54319227f50d8a77c751546534c804f
size 2643856128

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q6_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:4f55863d549aed46d95e712e47a8f39cf42c09f535b180d91d3983eedff287e7
size 2739278592

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q8_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:bf3ac06b1aaae8739d0d28802fb39023077856bb3283997a03c3e35eab3533c1
size 3421901568

3
NousResearch_DeepHermes-3-Llama-3-3B-Preview-bf16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:17c5ed1477cf84411e61849eb2f351d6d2fb1cdfb449d372d0dd851bea4b9f15
size 6433690080

BIN
NousResearch_DeepHermes-3-Llama-3-3B-Preview.imatrix Normal file
View File

Binary file not shown.

201
README.md Normal file
View File

@@ -0,0 +1,201 @@
---
quantized_by: bartowski
pipeline_tag: text-generation
widget:
- example_title: Hermes 3
messages:
- role: system
content: You are a sentient, superintelligent artificial general intelligence,
here to teach and assist me.
- role: user
content: What is the meaning of life?
language:
- en
tags:
- Llama-3
- instruct
- finetune
- chatml
- gpt4
- synthetic data
- distillation
- function calling
- json mode
- axolotl
- roleplaying
- chat
- reasoning
- r1
- vllm
license: llama3
base_model: NousResearch/DeepHermes-3-Llama-3-3B-Preview
model-index:
- name: DeepHermes-3-Llama-3.1-3B
results: []
---
## Llamacpp imatrix Quantizations of DeepHermes-3-Llama-3-3B-Preview by NousResearch
Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b4877">b4877</a> for quantization.
Original model: https://huggingface.co/NousResearch/DeepHermes-3-Llama-3-3B-Preview
All quants made using imatrix option with dataset from [here](https://gist.github.com/bartowski1182/eb213dccb3571f863da82e99418f81e8)
Run them in [LM Studio](https://lmstudio.ai/)
Run them directly with [llama.cpp](https://github.com/ggerganov/llama.cpp), or any other llama.cpp based project
## Prompt format
```
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
{system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|>
{prompt}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
```
## Download a file (not the whole branch) from below:
| Filename | Quant type | File Size | Split | Description |
| -------- | ---------- | --------- | ----- | ----------- |
| [DeepHermes-3-Llama-3-3B-Preview-bf16.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-bf16.gguf) | bf16 | 6.43GB | false | Full BF16 weights. |
| [DeepHermes-3-Llama-3-3B-Preview-Q8_0.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q8_0.gguf) | Q8_0 | 3.42GB | false | Extremely high quality, generally unneeded but max available quant. |
| [DeepHermes-3-Llama-3-3B-Preview-Q6_K_L.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q6_K_L.gguf) | Q6_K_L | 2.74GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
| [DeepHermes-3-Llama-3-3B-Preview-Q6_K.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q6_K.gguf) | Q6_K | 2.64GB | false | Very high quality, near perfect, *recommended*. |
| [DeepHermes-3-Llama-3-3B-Preview-Q5_K_L.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q5_K_L.gguf) | Q5_K_L | 2.42GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. |
| [DeepHermes-3-Llama-3-3B-Preview-Q5_K_M.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q5_K_M.gguf) | Q5_K_M | 2.32GB | false | High quality, *recommended*. |
| [DeepHermes-3-Llama-3-3B-Preview-Q5_K_S.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q5_K_S.gguf) | Q5_K_S | 2.27GB | false | High quality, *recommended*. |
| [DeepHermes-3-Llama-3-3B-Preview-Q4_K_L.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q4_K_L.gguf) | Q4_K_L | 2.11GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. |
| [DeepHermes-3-Llama-3-3B-Preview-Q4_1.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q4_1.gguf) | Q4_1 | 2.09GB | false | Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon. |
| [DeepHermes-3-Llama-3-3B-Preview-Q4_K_M.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q4_K_M.gguf) | Q4_K_M | 2.02GB | false | Good quality, default size for most use cases, *recommended*. |
| [DeepHermes-3-Llama-3-3B-Preview-Q4_K_S.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q4_K_S.gguf) | Q4_K_S | 1.93GB | false | Slightly lower quality with more space savings, *recommended*. |
| [DeepHermes-3-Llama-3-3B-Preview-Q4_0.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q4_0.gguf) | Q4_0 | 1.92GB | false | Legacy format, offers online repacking for ARM and AVX CPU inference. |
| [DeepHermes-3-Llama-3-3B-Preview-IQ4_NL.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-IQ4_NL.gguf) | IQ4_NL | 1.92GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. |
| [DeepHermes-3-Llama-3-3B-Preview-Q3_K_XL.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q3_K_XL.gguf) | Q3_K_XL | 1.91GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. |
| [DeepHermes-3-Llama-3-3B-Preview-IQ4_XS.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-IQ4_XS.gguf) | IQ4_XS | 1.83GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
| [DeepHermes-3-Llama-3-3B-Preview-Q3_K_L.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q3_K_L.gguf) | Q3_K_L | 1.82GB | false | Lower quality but usable, good for low RAM availability. |
| [DeepHermes-3-Llama-3-3B-Preview-Q3_K_M.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q3_K_M.gguf) | Q3_K_M | 1.69GB | false | Low quality. |
| [DeepHermes-3-Llama-3-3B-Preview-IQ3_M.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-IQ3_M.gguf) | IQ3_M | 1.60GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
| [DeepHermes-3-Llama-3-3B-Preview-Q3_K_S.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q3_K_S.gguf) | Q3_K_S | 1.54GB | false | Low quality, not recommended. |
| [DeepHermes-3-Llama-3-3B-Preview-IQ3_XS.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-IQ3_XS.gguf) | IQ3_XS | 1.48GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
| [DeepHermes-3-Llama-3-3B-Preview-Q2_K_L.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q2_K_L.gguf) | Q2_K_L | 1.46GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. |
| [DeepHermes-3-Llama-3-3B-Preview-Q2_K.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q2_K.gguf) | Q2_K | 1.36GB | false | Very low quality but surprisingly usable. |
| [DeepHermes-3-Llama-3-3B-Preview-IQ3_XXS.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-IQ3_XXS.gguf) | IQ3_XXS | 1.35GB | false | Lower quality, new method with decent performance, comparable to Q3 quants. |
| [DeepHermes-3-Llama-3-3B-Preview-IQ2_M.gguf](https://huggingface.co/bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF/blob/main/NousResearch_DeepHermes-3-Llama-3-3B-Preview-IQ2_M.gguf) | IQ2_M | 1.23GB | false | Relatively low quality, uses SOTA techniques to be surprisingly usable. |
## Embed/output weights
Some of these quants (Q3_K_XL, Q4_K_L etc) are the standard quantization method with the embeddings and output weights quantized to Q8_0 instead of what they would normally default to.
## Downloading using huggingface-cli
<details>
<summary>Click to view download instructions</summary>
First, make sure you have hugginface-cli installed:
```
pip install -U "huggingface_hub[cli]"
```
Then, you can target the specific file you want:
```
huggingface-cli download bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF --include "NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q4_K_M.gguf" --local-dir ./
```
If the model is bigger than 50GB, it will have been split into multiple files. In order to download them all to a local folder, run:
```
huggingface-cli download bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF --include "NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q8_0/*" --local-dir ./
```
You can either specify a new local-dir (NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q8_0) or download them all in place (./)
</details>
## ARM/AVX information
Previously, you would download Q4_0_4_4/4_8/8_8, and these would have their weights interleaved in memory in order to improve performance on ARM and AVX machines by loading up more data in one pass.
Now, however, there is something called "online repacking" for weights. details in [this PR](https://github.com/ggerganov/llama.cpp/pull/9921). If you use Q4_0 and your hardware would benefit from repacking weights, it will do it automatically on the fly.
As of llama.cpp build [b4282](https://github.com/ggerganov/llama.cpp/releases/tag/b4282) you will not be able to run the Q4_0_X_X files and will instead need to use Q4_0.
Additionally, if you want to get slightly better quality for , you can use IQ4_NL thanks to [this PR](https://github.com/ggerganov/llama.cpp/pull/10541) which will also repack the weights for ARM, though only the 4_4 for now. The loading time may be slower but it will result in an overall speed incrase.
<details>
<summary>Click to view Q4_0_X_X information (deprecated</summary>
I'm keeping this section to show the potential theoretical uplift in performance from using the Q4_0 with online repacking.
<details>
<summary>Click to view benchmarks on an AVX2 system (EPYC7702)</summary>
| model | size | params | backend | threads | test | t/s | % (vs Q4_0) |
| ------------------------------ | ---------: | ---------: | ---------- | ------: | ------------: | -------------------: |-------------: |
| qwen2 3B Q4_0 | 1.70 GiB | 3.09 B | CPU | 64 | pp512 | 204.03 ± 1.03 | 100% |
| qwen2 3B Q4_0 | 1.70 GiB | 3.09 B | CPU | 64 | pp1024 | 282.92 ± 0.19 | 100% |
| qwen2 3B Q4_0 | 1.70 GiB | 3.09 B | CPU | 64 | pp2048 | 259.49 ± 0.44 | 100% |
| qwen2 3B Q4_0 | 1.70 GiB | 3.09 B | CPU | 64 | tg128 | 39.12 ± 0.27 | 100% |
| qwen2 3B Q4_0 | 1.70 GiB | 3.09 B | CPU | 64 | tg256 | 39.31 ± 0.69 | 100% |
| qwen2 3B Q4_0 | 1.70 GiB | 3.09 B | CPU | 64 | tg512 | 40.52 ± 0.03 | 100% |
| qwen2 3B Q4_K_M | 1.79 GiB | 3.09 B | CPU | 64 | pp512 | 301.02 ± 1.74 | 147% |
| qwen2 3B Q4_K_M | 1.79 GiB | 3.09 B | CPU | 64 | pp1024 | 287.23 ± 0.20 | 101% |
| qwen2 3B Q4_K_M | 1.79 GiB | 3.09 B | CPU | 64 | pp2048 | 262.77 ± 1.81 | 101% |
| qwen2 3B Q4_K_M | 1.79 GiB | 3.09 B | CPU | 64 | tg128 | 18.80 ± 0.99 | 48% |
| qwen2 3B Q4_K_M | 1.79 GiB | 3.09 B | CPU | 64 | tg256 | 24.46 ± 3.04 | 83% |
| qwen2 3B Q4_K_M | 1.79 GiB | 3.09 B | CPU | 64 | tg512 | 36.32 ± 3.59 | 90% |
| qwen2 3B Q4_0_8_8 | 1.69 GiB | 3.09 B | CPU | 64 | pp512 | 271.71 ± 3.53 | 133% |
| qwen2 3B Q4_0_8_8 | 1.69 GiB | 3.09 B | CPU | 64 | pp1024 | 279.86 ± 45.63 | 100% |
| qwen2 3B Q4_0_8_8 | 1.69 GiB | 3.09 B | CPU | 64 | pp2048 | 320.77 ± 5.00 | 124% |
| qwen2 3B Q4_0_8_8 | 1.69 GiB | 3.09 B | CPU | 64 | tg128 | 43.51 ± 0.05 | 111% |
| qwen2 3B Q4_0_8_8 | 1.69 GiB | 3.09 B | CPU | 64 | tg256 | 43.35 ± 0.09 | 110% |
| qwen2 3B Q4_0_8_8 | 1.69 GiB | 3.09 B | CPU | 64 | tg512 | 42.60 ± 0.31 | 105% |
Q4_0_8_8 offers a nice bump to prompt processing and a small bump to text generation
</details>
</details>
## Which file should I choose?
<details>
<summary>Click here for details</summary>
A great write up with charts showing various performances is provided by Artefact2 [here](https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9)
The first thing to figure out is how big a model you can run. To do this, you'll need to figure out how much RAM and/or VRAM you have.
If you want your model running as FAST as possible, you'll want to fit the whole thing on your GPU's VRAM. Aim for a quant with a file size 1-2GB smaller than your GPU's total VRAM.
If you want the absolute maximum quality, add both your system RAM and your GPU's VRAM together, then similarly grab a quant with a file size 1-2GB Smaller than that total.
Next, you'll need to decide if you want to use an 'I-quant' or a 'K-quant'.
If you don't want to think too much, grab one of the K-quants. These are in format 'QX_K_X', like Q5_K_M.
If you want to get more into the weeds, you can check out this extremely useful feature chart:
[llama.cpp feature matrix](https://github.com/ggerganov/llama.cpp/wiki/Feature-matrix)
But basically, if you're aiming for below Q4, and you're running cuBLAS (Nvidia) or rocBLAS (AMD), you should look towards the I-quants. These are in format IQX_X, like IQ3_M. These are newer and offer better performance for their size.
These I-quants can also be used on CPU, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.
The I-quants are *not* compatible with Vulcan, which is also AMD, so if you have an AMD card double check if you're using the rocBLAS build or the Vulcan build. At the time of writing this, LM Studio has a preview with ROCm support, and other inference engines have specific builds for ROCm.
</details>
## Credits
Thank you kalomaze and Dampf for assistance in creating the imatrix calibration dataset.
Thank you ZeroWw for the inspiration to experiment with embed/output.
Thank you to LM Studio for sponsoring my work.
Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski

1
configuration.json Normal file
View File

@@ -0,0 +1 @@
{"framework": "pytorch", "task": "text-generation", "allow_remote": true}