初始化项目,由ModelHub XC社区提供模型

Model: bartowski/NaturalLM-7B-Instruct-GGUF
Source: Original Platform
This commit is contained in:
ModelHub XC committed 2026-11-05 06:42:12 +08:00
commit 966044ce0f
29 files changed
+317

No files matched your search

+61
View File
@@ -0,0 +1,61 @@
*.7z filter=lfs diff=lfs merge=lfs -text
*.arrow filter=lfs diff=lfs merge=lfs -text
*.bin filter=lfs diff=lfs merge=lfs -text
*.bz2 filter=lfs diff=lfs merge=lfs -text
*.ckpt 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
*.mlmodel filter=lfs diff=lfs merge=lfs -text
*.model filter=lfs diff=lfs merge=lfs -text
*.msgpack filter=lfs diff=lfs merge=lfs -text
*.npy filter=lfs diff=lfs merge=lfs -text
*.npz 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
*.pickle filter=lfs diff=lfs merge=lfs -text
*.pkl 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
*.safetensors filter=lfs diff=lfs merge=lfs -text
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
*.tar.* 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
*.wasm filter=lfs diff=lfs merge=lfs -text
*.xz filter=lfs diff=lfs merge=lfs -text
*.zip filter=lfs diff=lfs merge=lfs -text
*.zst filter=lfs diff=lfs merge=lfs -text
*tfevents* filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q6_K_L.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q5_K_L.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q4_K_L.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q4_0_8_8.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q4_0_4_8.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q4_0_4_4.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-IQ4_NL.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q3_K_XL.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-IQ3_XS.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q2_K_L.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-IQ2_M.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct-f16.gguf filter=lfs diff=lfs merge=lfs -text
NaturalLM-7B-Instruct.imatrix filter=lfs diff=lfs merge=lfs -text
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d8a5315368f840f3ab77de71836fcca4554651a99d15a57168b0767f85f2da1d
size 2504250464
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:627ed0d9a6310a03b53fdb2ee3ab3595f5196eb5bdaf95f30d43996899e8e9e8
size 3288847456
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:9949b40607d5f42ec577450b6907a109bf4515a9c41cdebff0bd0c86d3a45a3e
size 3022771296
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b6992704028bc7119116b80a922e79e48ab9de1cfb9082597482674bf2f3af14
size 4130067552
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:4ac6a8c3a78bb2253dcd3cee7fa947cb284612cbd1795ebfa0a5702768a65605
size 3911963744
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:370aad7121fa60f0880a7c5aac430fe474284ed00db5484bc7c8ac7cbc7095a6
size 2722878560
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:28acb0cc73b9ecfc9ea7c8f87356a0a4e27814e85ef19c898eb3478e07d7a9a1
size 2853950560
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:9b35e4b248422cee68da59ada212487b6fa072f22debc8950e37b5175b09b564
size 3825980512
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3cfe91ad3aafac1aca2e2257ccd535c9c69002d8a081c5a0ab16ecd64a726a71
size 3522942048
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:6ee4eb3afc52b4f924f2c4280fb1b6092d817636fa1919359c5de95f45774742
size 3168523360
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ebd023a8e503ce9348b615e3c4abda1b05faa33445be1848291210056d44ce73
size 3943421024
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:4b9fae21e6bd0a3b034edaec8073132f5a060ebb10b4968836b36fb22203e029
size 4127970400
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3bc0ef6181aa7a78dc62fa8e527e905ca7fb87a1a3ac13208da67cd297a3bab2
size 4113290336
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:6d7aa70323d474ed6eac25cfa165f57cfb32b3c5ef2faa1a60c71ea549c09446
size 4113290336
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3f36d5b7575e54343175030e411f089b616670871edd2d18db08a6218dd48249
size 4113290336
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b021276f80303bd5e45da18942c534487cedf0a513522ef097546bcd58ea45e5
size 4472427616
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:15b2f34116f690fea35790a9392b8a2190fe25827e370d426e88a2a543f4dcee
size 4372812896
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b35075d0af6d899f81b3df1d2cd5134abc3524abe7ba5256c2d2a4db686aa929
size 4144747616
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:0e8935c21dfef6b217fe4f2ce78714d0e01f09e9263cee372f089a1d703650f7
size 5219013728
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:95245e2c2fe9cf98631d2b7642aa3318666d4992aecb4172b7676589323ddae4
size 5136176224
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:7d2a68794652a6f41e93f734cf870fe3e965914ce8b38b0b9c7d09308c79a605
size 5002482784
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:e0402590882a2cff090b16c50e2936c14cbe131009063e95e738a2f5009dd0db
size 5947249760
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:5827b6169d48239400184f7b1cb4575e6d9ad0985018936c2c4537a2f9a5998a
size 6012261472
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b150a9f5cab079df1cf94675ed1540528644f658dd02ebc549fb765baf783aa6
size 7702565984
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:49eaa8487942c882457b80efe53f0740018ee49cd7779788940bf85d7ed12b5d
size 14497338176
+3
View File
@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:717fe800e08079e3c2787034d4fbae1f93be7ae5961ebc4ed3353223b9b48f6b
size 4988170
+177
View File
@@ -0,0 +1,177 @@
---
quantized_by: bartowski
pipeline_tag: text-generation
base_model: qingy2024/NaturalLM-7B-Instruct
tags:
- unsloth
- trl
- sft
---
## Llamacpp imatrix Quantizations of NaturalLM-7B-Instruct
Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b4273">b4273</a> for quantization.
Original model: https://huggingface.co/qingy2024/NaturalLM-7B-Instruct
All quants made using imatrix option with dataset from [here](https://gist.github.com/bartowski1182/eb213dccb3571f863da82e99418f81e8)
Run them in [LM Studio](https://lmstudio.ai/)
## Prompt format
```
<|im_start|>system
{system_prompt}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
```
## Download a file (not the whole branch) from below:
| Filename | Quant type | File Size | Split | Description |
| -------- | ---------- | --------- | ----- | ----------- |
| [NaturalLM-7B-Instruct-f16.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-f16.gguf) | f16 | 14.50GB | false | Full F16 weights. |
| [NaturalLM-7B-Instruct-Q8_0.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q8_0.gguf) | Q8_0 | 7.70GB | false | Extremely high quality, generally unneeded but max available quant. |
| [NaturalLM-7B-Instruct-Q6_K_L.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q6_K_L.gguf) | Q6_K_L | 6.01GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
| [NaturalLM-7B-Instruct-Q6_K.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q6_K.gguf) | Q6_K | 5.95GB | false | Very high quality, near perfect, *recommended*. |
| [NaturalLM-7B-Instruct-Q5_K_L.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q5_K_L.gguf) | Q5_K_L | 5.22GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. |
| [NaturalLM-7B-Instruct-Q5_K_M.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q5_K_M.gguf) | Q5_K_M | 5.14GB | false | High quality, *recommended*. |
| [NaturalLM-7B-Instruct-Q5_K_S.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q5_K_S.gguf) | Q5_K_S | 5.00GB | false | High quality, *recommended*. |
| [NaturalLM-7B-Instruct-Q4_K_L.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q4_K_L.gguf) | Q4_K_L | 4.47GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. |
| [NaturalLM-7B-Instruct-Q4_K_M.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q4_K_M.gguf) | Q4_K_M | 4.37GB | false | Good quality, default size for most use cases, *recommended*. |
| [NaturalLM-7B-Instruct-Q4_K_S.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q4_K_S.gguf) | Q4_K_S | 4.14GB | false | Slightly lower quality with more space savings, *recommended*. |
| [NaturalLM-7B-Instruct-Q4_0.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q4_0.gguf) | Q4_0 | 4.13GB | false | Legacy format, offers online repacking for ARM CPU inference. |
| [NaturalLM-7B-Instruct-IQ4_NL.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-IQ4_NL.gguf) | IQ4_NL | 4.13GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. |
| [NaturalLM-7B-Instruct-Q4_0_8_8.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q4_0_8_8.gguf) | Q4_0_8_8 | 4.11GB | false | Optimized for ARM and AVX inference. Requires 'sve' support for ARM (see details below). *Don't use on Mac*. |
| [NaturalLM-7B-Instruct-Q4_0_4_8.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q4_0_4_8.gguf) | Q4_0_4_8 | 4.11GB | false | Optimized for ARM inference. Requires 'i8mm' support (see details below). *Don't use on Mac*. |
| [NaturalLM-7B-Instruct-Q4_0_4_4.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q4_0_4_4.gguf) | Q4_0_4_4 | 4.11GB | false | Optimized for ARM inference. Should work well on all ARM chips, not for use with GPUs. *Don't use on Mac*. |
| [NaturalLM-7B-Instruct-Q3_K_XL.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q3_K_XL.gguf) | Q3_K_XL | 3.94GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. |
| [NaturalLM-7B-Instruct-IQ4_XS.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-IQ4_XS.gguf) | IQ4_XS | 3.91GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
| [NaturalLM-7B-Instruct-Q3_K_L.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q3_K_L.gguf) | Q3_K_L | 3.83GB | false | Lower quality but usable, good for low RAM availability. |
| [NaturalLM-7B-Instruct-Q3_K_M.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q3_K_M.gguf) | Q3_K_M | 3.52GB | false | Low quality. |
| [NaturalLM-7B-Instruct-IQ3_M.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-IQ3_M.gguf) | IQ3_M | 3.29GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
| [NaturalLM-7B-Instruct-Q3_K_S.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q3_K_S.gguf) | Q3_K_S | 3.17GB | false | Low quality, not recommended. |
| [NaturalLM-7B-Instruct-IQ3_XS.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-IQ3_XS.gguf) | IQ3_XS | 3.02GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
| [NaturalLM-7B-Instruct-Q2_K_L.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q2_K_L.gguf) | Q2_K_L | 2.85GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. |
| [NaturalLM-7B-Instruct-Q2_K.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-Q2_K.gguf) | Q2_K | 2.72GB | false | Very low quality but surprisingly usable. |
| [NaturalLM-7B-Instruct-IQ2_M.gguf](https://huggingface.co/bartowski/NaturalLM-7B-Instruct-GGUF/blob/main/NaturalLM-7B-Instruct-IQ2_M.gguf) | IQ2_M | 2.50GB | 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/NaturalLM-7B-Instruct-GGUF --include "NaturalLM-7B-Instruct-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/NaturalLM-7B-Instruct-GGUF --include "NaturalLM-7B-Instruct-Q8_0/*" --local-dir ./
```
You can either specify a new local-dir (NaturalLM-7B-Instruct-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</summary>
These are *NOT* for Metal (Apple) or GPU (nvidia/AMD/intel) offloading, only ARM chips (and certain AVX2/AVX512 CPUs).
If you're using an ARM chip, the Q4_0_X_X quants will have a substantial speedup. Check out Q4_0_4_4 speed comparisons [on the original pull request](https://github.com/ggerganov/llama.cpp/pull/5780#pullrequestreview-21657544660)
To check which one would work best for your ARM chip, you can check [AArch64 SoC features](https://gpages.juszkiewicz.com.pl/arm-socs-table/arm-socs.html) (thanks EloyOn!).
If you're using a CPU that supports AVX2 or AVX512 (typically server CPUs and AMD's latest Zen5 CPUs) and are not offloading to a GPU, the Q4_0_8_8 may offer a nice speed as well:
<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 and Apple Metal, 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.
Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski
+1
View File
@@ -0,0 +1 @@
{"framework": "pytorch", "task": "text-generation", "allow_remote": true}