Upload README.md
Browse files
README.md
CHANGED
@@ -98,12 +98,12 @@ Most GPTQ files are made with AutoGPTQ. Mistral models are currently made with T
|
|
98 |
|
99 |
| Branch | Bits | GS | Act Order | Damp % | GPTQ Dataset | Seq Len | Size | ExLlama | Desc |
|
100 |
| ------ | ---- | -- | --------- | ------ | ------------ | ------- | ---- | ------- | ---- |
|
101 |
-
| main | 4 | 128 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) |
|
102 |
-
| gptq-4bit-32g-actorder_True | 4 | 32 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) |
|
103 |
-
| gptq-8bit--1g-actorder_True | 8 | None | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) |
|
104 |
-
| gptq-8bit-128g-actorder_True | 8 | 128 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) |
|
105 |
-
| gptq-8bit-32g-actorder_True | 8 | 32 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) |
|
106 |
-
| gptq-4bit-64g-actorder_True | 4 | 64 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) |
|
107 |
|
108 |
<!-- README_GPTQ.md-provided-files end -->
|
109 |
|
|
|
98 |
|
99 |
| Branch | Bits | GS | Act Order | Damp % | GPTQ Dataset | Seq Len | Size | ExLlama | Desc |
|
100 |
| ------ | ---- | -- | --------- | ------ | ------------ | ------- | ---- | ------- | ---- |
|
101 |
+
| main | 4 | 128 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 0.23 GB | No | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. |
|
102 |
+
| gptq-4bit-32g-actorder_True | 4 | 32 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 0.24 GB | No | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. |
|
103 |
+
| gptq-8bit--1g-actorder_True | 8 | None | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 0.31 GB | No | 8-bit, with Act Order. No group size, to lower VRAM requirements. |
|
104 |
+
| gptq-8bit-128g-actorder_True | 8 | 128 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 0.32 GB | No | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. |
|
105 |
+
| gptq-8bit-32g-actorder_True | 8 | 32 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 0.33 GB | No | 8-bit, with group size 32g and Act Order for maximum inference quality. |
|
106 |
+
| gptq-4bit-64g-actorder_True | 4 | 64 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 0.23 GB | No | 4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy. |
|
107 |
|
108 |
<!-- README_GPTQ.md-provided-files end -->
|
109 |
|