nwdxlgzs/XL-LuaCopilot-1.7B-FFT-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: nwdxlgzs/XL-LuaCopilot-1.7B-FFT-GGUF
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-08-29 20:28:17 +08:00
commit a05a0c1799
26 changed files with 272 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

59
.gitattributes vendored Normal file
View File

@@ -0,0 +1,59 @@
*.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
XL-LuaCopilot-1.7B-FFT-BF16.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-F16.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-IQ3_S.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-IQ3_XS.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-IQ3_XXS.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-IQ4_NL.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q3_K.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q4_1.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q4_K.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q5_0.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q5_1.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q5_K.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
XL-LuaCopilot-1.7B-FFT-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text

141
README.md Normal file
View File

@@ -0,0 +1,141 @@
---
tags:
- unsloth
- lua
base_model:
- nwdxlgzs/XL-LuaCopilot-1.7B-FFT-checkpoint-46000
license: gpl-3.0
library_name: transformers
pipeline_tag: text-generation
---
# XL-LuaCopilot-1.7B-FFT
XL-LuaCopilot-1.7B-FFT is a large language model (LLM) based on the Qwen architecture(Qwen3-1.7B), specifically designed for code generation tasks in Lua programming language. It has been full fine-tuned (FFT) to improve its performance and efficiency when generating Lua code.
I sugggest you use `"chat_template_kwargs": {"enable_thinking": false}` because my train data with none thinking. I also found low `temperature` ususually works well for code generation tasks.
quantize=["Q4_0", "Q4_1", "Q5_0", "Q5_1", "IQ3_XXS", "IQ3_S", "IQ3_M", "Q3_K", "IQ3_XS", "Q3_K_S", "Q3_K_M", "Q3_K_L", "IQ4_NL", "IQ4_XS", "Q4_K", "Q4_K_S", "Q4_K_M", "Q5_K", "Q5_K_S", "Q5_K_M", "Q6_K", "Q8_0", "F16", "BF16"]
> `checkpoint-37000` is the checkpoint where the model had just entered the plateau phase with a lower loss, and it might be better than `checkpoint-46000`. However, the GGUF files I provide will still be released based on the final checkpoint at the end of training.
## Train Samples
1472000 (steps=46000)x(per_device_train_batch_size=8)x(gradient_accumulation_steps=4)x(device=1)
datasets: 1464339 (luafiles=488113)x(split=3)
epoch‌s = 1.005
## How To Use
> With OpenAI Compatible API (llama.cpp:llama-server)
```json
-> REQUEST ->
{
"model": "XL-LuaCopilot-1.7B-FFT",
"messages": [
{"role": "system","content": "prefix"},
{"role": "user","content": "do\n--打印：你好世界\n local tex"},
{"role": "system","content": "suffix"},
{"role": "user","content": "nd"},
{"role": "system","content": "middle"}
],
"stream": false,
"cache_prompt": false,
"samplers": "edkypmxt",
"temperature": 0.2,
"dynatemp_range": 0.1,
"dynatemp_exponent": 1,
"top_k": 20,
"top_p": 0.9,
"min_p": 0.05,
"typical_p": 1,
"xtc_probability": 0,
"xtc_threshold": 0.1,
"repeat_last_n": 32,
"repeat_penalty": 1.1,
"presence_penalty": 0,
"frequency_penalty": 0.5,
"dry_multiplier": 0,
"dry_base": 1.75,
"dry_allowed_length": 2,
"dry_penalty_last_n": -1,
"max_tokens": -1,
"timings_per_token": true,
"chat_template_kwargs": {"enable_thinking": false}
}
-> RESPONSE ->
{
"choices": [
{
"finish_reason": "stop",
"index": 0,
"message": {
"role": "assistant",
"content": "<think>\n\n</think>\n\nt = \"你好世界\"\n print(text)\ne"
}
}
],
...
}
```
> I know Qwen has `<|fim_prefix|>` / `<|fim_suffix|>` / `<|fim_middle|>` tokens, but I'm not sure Qwen3 trains these tokens (I just know Qwen2.5-Coder does). To use code generation easily, I use chatml format.
> If you just want to chat with it, you can use some tricks like this:
```
<|im_end|>
<|im_start|>system
prefix<|im_end|>
<|im_start|>user
do
--打印：你好世界
local tex<|im_end|>
<|im_start|>system
suffix<|im_end|>
<|im_start|>user
nd<|im_end|>
<|im_start|>system
middle
```
It dosen't work very well, but it's a good way let you fast try. It will convert to this prompt text:
```
<|im_start|>user
<|im_end|>
<|im_start|>system
prefix<|im_end|>
<|im_start|>user
do
--打印：你好世界
local tex<|im_end|>
<|im_start|>system
suffix<|im_end|>
<|im_start|>user
nd<|im_end|>
<|im_start|>system
middle<|im_end|>
```
Hope model skip first `<|im_start|>user\n<|im_end|>` part.
# Train Device
> Online GPU is Expensive !
| 类别 | 配置详情 |
|----------------|---------------------------------------------------|
| **镜像** | Ubuntu 22.04 |
| **PyTorch** | 2.5.1 |
| **Python** | 3.12 |
| **CUDA** | 12.4 |
| **GPU** | RTX 4090 (24GB) * 1 |
| **CPU** | 25 vCPU Intel(R) Xeon(R) Platinum 8481C |
| **内存** | 90GB |
| **硬盘** | 30 GB + 50 GB |
| **时长** | 3 Day |

3
XL-LuaCopilot-1.7B-FFT-BF16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:680d40cbecc8177f2901be94c845bd0bf578b670d9f5250ef38f4d72e2c2ada1
size 3447344704

3
XL-LuaCopilot-1.7B-FFT-F16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d07e8a926af4d363f515ccb55b875852d151ccd00b5b04452f7c6bad5e74a7eb
size 3447344704

3
XL-LuaCopilot-1.7B-FFT-IQ3_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:2d33d7a07947b2e6ad28d1e8cfddeb7dc6131c700e214f85e12549816a149185
size 895657536

3
XL-LuaCopilot-1.7B-FFT-IQ3_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:bda2e01db0a5db5c22150677a16de6b477c40fb4ad3b063245ac493711520b6f
size 867247680

3
XL-LuaCopilot-1.7B-FFT-IQ3_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:0e5532034d76d5607be3ea39d841c3e063487409f2f3d91aa70e6c9647dee058
size 834217536

3
XL-LuaCopilot-1.7B-FFT-IQ3_XXS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:9c7687d47e32f149f8656a0e247a1dd11f83eadbd12337cb391eda3aa132531b
size 219881024

3
XL-LuaCopilot-1.7B-FFT-IQ4_NL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:1dacc5b6868ccb2cc7a856acb5e3a1b100736c1736777e6c27f75a45713c4eba
size 1059137088

3
XL-LuaCopilot-1.7B-FFT-IQ4_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:48024c6a36ceb2ef2289a78d7a144f781acb459bde30842b1614655f8bc8d530
size 1016276544

3
XL-LuaCopilot-1.7B-FFT-Q3_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3d06ab461534367d70e6161938ae3e173e05db61ca3ec19e048743def920aa1a
size 939533888

3
XL-LuaCopilot-1.7B-FFT-Q3_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:671f230537a1c9b9020ce02e2a2f515d0092368f17560b4314a662cc6b0ae3bd
size 1003497024

3
XL-LuaCopilot-1.7B-FFT-Q3_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3d06ab461534367d70e6161938ae3e173e05db61ca3ec19e048743def920aa1a
size 939533888

3
XL-LuaCopilot-1.7B-FFT-Q3_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ea5448930564e5a6f3daca95ac2a28fab61dd77db8d72f019ebd511eaa5cc556
size 867247680

3
XL-LuaCopilot-1.7B-FFT-Q4_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:722bb4e698ce94eca1c7bd4b6ed9b5d4562426783e5336c24a65cc46f292cae1
size 1054418496

3
XL-LuaCopilot-1.7B-FFT-Q4_1.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:a1a32b5088a56f37ac8af7db6f1ad7f393e3949dfd717e4943ccefb4a5de36c5
size 1142498880

3
XL-LuaCopilot-1.7B-FFT-Q4_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:7379ec6d98eb94408d63db3cc5ef5d7b9d7f63e380177a3ca6d8259184d8d298
size 1107404352

3
XL-LuaCopilot-1.7B-FFT-Q4_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:7379ec6d98eb94408d63db3cc5ef5d7b9d7f63e380177a3ca6d8259184d8d298
size 1107404352

3
XL-LuaCopilot-1.7B-FFT-Q4_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:9198790fa8d492990467f615ce906e8e1ad5fe881c4d951e4f84c0b715f94716
size 1060185664

3
XL-LuaCopilot-1.7B-FFT-Q5_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:a818d8d6f0f798c18638a2390c77663e3c6c9db6ecb0171c54ebb265523bf617
size 1230579264

3
XL-LuaCopilot-1.7B-FFT-Q5_1.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b57452e2ba7043e6299248c134a206907eb4f0107a2ad2405b8d1dbd3d6b24ec
size 1318659648

3
XL-LuaCopilot-1.7B-FFT-Q5_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:a5d5735b9575c18cfcc2ed851ac6dd76f82e88704d852020f238ac390578fa8c
size 1257875008

3
XL-LuaCopilot-1.7B-FFT-Q5_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:a5d5735b9575c18cfcc2ed851ac6dd76f82e88704d852020f238ac390578fa8c
size 1257875008

3
XL-LuaCopilot-1.7B-FFT-Q5_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:4707bd98d25f78bc5f9cece5ae3a296dda6588d10da57026a8ed43b298130f5e
size 1230579264

3
XL-LuaCopilot-1.7B-FFT-Q6_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:46f5ef40d7083057500dcb66c2b454957bf2e1d83f25e431331b02d2538118ef
size 1417750080

3
XL-LuaCopilot-1.7B-FFT-Q8_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:de830da67aee50ae76268213b3e00375c8cde6180d1b0b089cb26e67e50715dd
size 1834421824