bartowski/Falcon3-1B-Instruct-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/Falcon3-1B-Instruct-GGUF
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-06-29 14:22:12 +08:00
commit b9ec1daf6f
26 changed files with 307 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

58
.gitattributes vendored Normal file
View File

@@ -0,0 +1,58 @@
*.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
Falcon3-1B-Instruct-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-Q6_K_L.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-Q5_K_L.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-Q4_K_L.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-IQ4_NL.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-Q3_K_XL.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-IQ3_XS.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-Q2_K_L.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-IQ2_M.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct-f16.gguf filter=lfs diff=lfs merge=lfs -text
Falcon3-1B-Instruct.imatrix filter=lfs diff=lfs merge=lfs -text

3
Falcon3-1B-Instruct-IQ2_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:02e8aab64718ec8d63320c06b7caa83d51e4ce7e1fbd6b95d2c734d5a374ece0
size 683735200

3
Falcon3-1B-Instruct-IQ3_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:945b99e76af4f10f35f13d43f1d3ad5884892faff8c21b3d28761c6d9f42fbf2
size 846690464

3
Falcon3-1B-Instruct-IQ3_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:371d6b0640f3fad2ddb0fa14b2de97d05b00de772c7f388eba40b4d464d3e0fb
size 801437856

3
Falcon3-1B-Instruct-IQ4_NL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ff0a67a7132c75dc90127dcf8c3e5a4aff2a4580138d77efd7c9351a1abbd432
size 1013250208

3
Falcon3-1B-Instruct-IQ4_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:627e49167468a178f7686dc848013ae751f219776c2b1bff6223a03a55756882
size 969472160

3
Falcon3-1B-Instruct-Q2_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:985ec25ec9ae922922a413e839002353b33685a6955a58cc8d006cfcbdc3aa22
size 727087264

3
Falcon3-1B-Instruct-Q2_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:0dbcd8151f4a4eccf35a9e7b9f272c216bbb5a1025e92acaf64177e556be95ad
size 989231264

3
Falcon3-1B-Instruct-Q3_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:5ba92e17c6f965437764ae4dda5d2607a8c6b624c094afdb1b9c75fe20b19d67
size 934246560

3
Falcon3-1B-Instruct-Q3_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d55d856a5a4efeaf60ab8ffdf18c26a2d3fe15e784e8ae0bee97aaf078bbc442
size 884963488

3
Falcon3-1B-Instruct-Q3_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ee5ec82dfea9b0273e5208191b5234c820b8f0b94bc95578057c6f3c37118fcd
size 827193504

3
Falcon3-1B-Instruct-Q3_K_XL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:75a456d02533635e1b68ccf769e73e28c974d642af3bcff9326a07f2754ace85
size 1169127584

3
Falcon3-1B-Instruct-Q4_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d36b33cd1f32df20f1e1f193061d70a16f174dec175edf40312c52f8ad4c9196
size 1015347360

3
Falcon3-1B-Instruct-Q4_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:db7535cad0f8a212ee89a818d06b08ca218cb5c44af73675601f546a82f398ca
size 1256274080

3
Falcon3-1B-Instruct-Q4_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:1c92013dac1ab6e703e787f3e0829ca03cc95311e4c113a77950d15ff6dea7b3
size 1057044640

3
Falcon3-1B-Instruct-Q4_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:76b5f16770a08e90cb98c06fdec26d7d22cdf831e92481660ca21ce4658f84fb
size 1018493088

3
Falcon3-1B-Instruct-Q5_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d3f7b4132a71c492c6fc7eb13c0c8ba9af352b02904e5746a40a78dfafe1f7a0
size 1376598176

3
Falcon3-1B-Instruct-Q5_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d02f0870e4bd8c14c20e62b7d59de55b5b5593a1ce34837fca0d4bce5f7575ff
size 1210923168

3
Falcon3-1B-Instruct-Q5_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:27b7896b40c5b4884c2c9eb299fddc6aceb3bff0ef912cc6d50e48b2ad39f139
size 1188362400

3
Falcon3-1B-Instruct-Q6_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:fd250417840e63299ecc622395d2793945730d1b47a1510b8c77d4b7576107c5
size 1374419104

3
Falcon3-1B-Instruct-Q6_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:f9797f63f6ad31d77c67dc2d0517e2e1574d7f0c27436a7e16be557512eece86
size 1504442528

3
Falcon3-1B-Instruct-Q8_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:50edacc1e8fe64b5449ff8ae76a0ed3a168de3ec7b49683d41c4b3bb48575a48
size 1778710688

3
Falcon3-1B-Instruct-f16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:2807c4440778122985679707d21759fe63d1212983800606cae8deb2cb449f0d
size 3343710080

3
Falcon3-1B-Instruct.imatrix Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c4467113475d9e96e6df511c6438b80edf1cb2fb18b9ea9e26dcf68d4a2c7730
size 1478732

179
README.md Normal file
View File

@@ -0,0 +1,179 @@
---
quantized_by: bartowski
pipeline_tag: text-generation
language:
- en
- fr
- es
- pt
license_name: falcon-llm-license
tags:
- falcon3
license_link: https://falconllm.tii.ae/falcon-terms-and-conditions.html
license: other
base_model: tiiuae/Falcon3-1B-Instruct
---
## Llamacpp imatrix Quantizations of Falcon3-1B-Instruct
Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b4381">b4381</a> for quantization.
Original model: https://huggingface.co/tiiuae/Falcon3-1B-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
```
<|system|>
{system_prompt}
<|user|>
{prompt}
<|assistant|>
```
## What's new:
Fix tokenizer
## Download a file (not the whole branch) from below:
| Filename | Quant type | File Size | Split | Description |
| -------- | ---------- | --------- | ----- | ----------- |
| [Falcon3-1B-Instruct-f16.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-f16.gguf) | f16 | 3.34GB | false | Full F16 weights. |
| [Falcon3-1B-Instruct-Q8_0.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q8_0.gguf) | Q8_0 | 1.78GB | false | Extremely high quality, generally unneeded but max available quant. |
| [Falcon3-1B-Instruct-Q6_K_L.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q6_K_L.gguf) | Q6_K_L | 1.50GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
| [Falcon3-1B-Instruct-Q5_K_L.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q5_K_L.gguf) | Q5_K_L | 1.38GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. |
| [Falcon3-1B-Instruct-Q6_K.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q6_K.gguf) | Q6_K | 1.37GB | false | Very high quality, near perfect, *recommended*. |
| [Falcon3-1B-Instruct-Q4_K_L.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q4_K_L.gguf) | Q4_K_L | 1.26GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. |
| [Falcon3-1B-Instruct-Q5_K_M.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q5_K_M.gguf) | Q5_K_M | 1.21GB | false | High quality, *recommended*. |
| [Falcon3-1B-Instruct-Q5_K_S.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q5_K_S.gguf) | Q5_K_S | 1.19GB | false | High quality, *recommended*. |
| [Falcon3-1B-Instruct-Q3_K_XL.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q3_K_XL.gguf) | Q3_K_XL | 1.17GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. |
| [Falcon3-1B-Instruct-Q4_K_M.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q4_K_M.gguf) | Q4_K_M | 1.06GB | false | Good quality, default size for most use cases, *recommended*. |
| [Falcon3-1B-Instruct-Q4_K_S.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q4_K_S.gguf) | Q4_K_S | 1.02GB | false | Slightly lower quality with more space savings, *recommended*. |
| [Falcon3-1B-Instruct-Q4_0.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q4_0.gguf) | Q4_0 | 1.02GB | false | Legacy format, offers online repacking for ARM and AVX CPU inference. |
| [Falcon3-1B-Instruct-IQ4_NL.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-IQ4_NL.gguf) | IQ4_NL | 1.01GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. |
| [Falcon3-1B-Instruct-Q2_K_L.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q2_K_L.gguf) | Q2_K_L | 0.99GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. |
| [Falcon3-1B-Instruct-IQ4_XS.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-IQ4_XS.gguf) | IQ4_XS | 0.97GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
| [Falcon3-1B-Instruct-Q3_K_L.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q3_K_L.gguf) | Q3_K_L | 0.93GB | false | Lower quality but usable, good for low RAM availability. |
| [Falcon3-1B-Instruct-Q3_K_M.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q3_K_M.gguf) | Q3_K_M | 0.88GB | false | Low quality. |
| [Falcon3-1B-Instruct-IQ3_M.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-IQ3_M.gguf) | IQ3_M | 0.85GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
| [Falcon3-1B-Instruct-Q3_K_S.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q3_K_S.gguf) | Q3_K_S | 0.83GB | false | Low quality, not recommended. |
| [Falcon3-1B-Instruct-IQ3_XS.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-IQ3_XS.gguf) | IQ3_XS | 0.80GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
| [Falcon3-1B-Instruct-Q2_K.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-Q2_K.gguf) | Q2_K | 0.73GB | false | Very low quality but surprisingly usable. |
| [Falcon3-1B-Instruct-IQ2_M.gguf](https://huggingface.co/bartowski/Falcon3-1B-Instruct-GGUF/blob/main/Falcon3-1B-Instruct-IQ2_M.gguf) | IQ2_M | 0.68GB | 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/Falcon3-1B-Instruct-GGUF --include "Falcon3-1B-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/Falcon3-1B-Instruct-GGUF --include "Falcon3-1B-Instruct-Q8_0/*" --local-dir ./
```
You can either specify a new local-dir (Falcon3-1B-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 (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 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
configuration.json Normal file
View File

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