Update README.md
Browse files
README.md
CHANGED
|
@@ -47,13 +47,16 @@ We introduce [SeaLLM-7B-v2.5](https://huggingface.co/SeaLLMs/SeaLLM-7B-v2.5), th
|
|
| 47 |
- DEMO: [SeaLLMs/SeaLLM-7B](https://huggingface.co/spaces/SeaLLMs/SeaLLM-7B).
|
| 48 |
- Technical report: [Arxiv: SeaLLMs - Large Language Models for Southeast Asia](https://arxiv.org/pdf/2312.00738.pdf).
|
| 49 |
- Model weights:
|
| 50 |
-
- [SeaLLM-7B-v2](https://huggingface.co/SeaLLMs/SeaLLM-7B-v2).
|
| 51 |
-
- [SeaLLM-7B-v2-GGUF](https://huggingface.co/SeaLLMs/SeaLLM-7B-v2.5-GGUF).
|
| 52 |
- Run locally:
|
| 53 |
- [LM-studio](https://lmstudio.ai/):
|
| 54 |
- [SeaLLM-7B-v2.5-q4_0-chatml](https://huggingface.co/SeaLLMs/SeaLLM-7B-v2.5-GGUF/blob/main/seallm-7b-v2.5-chatml.Q4_K_M.gguf) with ChatML template (`<eos>` token changed to `<|im_end|>`)
|
| 55 |
- [SeaLLM-7B-v2.5-q4_0](https://huggingface.co/SeaLLMs/SeaLLM-7B-v2.5-GGUF/blob/main/seallm-7b-v2.5.Q4_K_M.gguf) - must use SeaLLM-7B-v2.5 chat format.
|
| 56 |
- [MLX for Apple Silicon](https://github.com/ml-explore/mlx): [SeaLLMs/SeaLLM-7B-v2.5-mlx-quantized](https://huggingface.co/SeaLLMs/SeaLLM-7B-v2.5-mlx-quantized)
|
|
|
|
|
|
|
|
|
|
| 57 |
|
| 58 |
<blockquote style="color:red">
|
| 59 |
<p><strong style="color: red">Terms of Use and License</strong>:
|
|
@@ -109,7 +112,6 @@ We evaluate models on 3 benchmarks following the recommended default setups: 5-s
|
|
| 109 |
| SeaLLM-7B-v2.5 | 78.5 | 34.9 | 51.3 | 22.1 | 72.3 | 30.2 | 71.5 | 30.1 | 62.0 | 28.4
|
| 110 |
|
| 111 |
|
| 112 |
-
|
| 113 |
Baselines were evaluated using their respective chat-template and system prompts ([Qwen1.5-7B-chat](https://huggingface.co/Qwen/Qwen1.5-7B-Chat/blob/main/tokenizer_config.json), [Vistral](https://huggingface.co/Viet-Mistral/Vistral-7B-Chat)).
|
| 114 |
|
| 115 |
#### Zero-shot MGSM
|
|
@@ -124,7 +126,6 @@ Baselines were evaluated using their respective chat-template and system prompts
|
|
| 124 |
| SeaLLM-7B-v2.5 | 58.0 | **64.8**
|
| 125 |
|
| 126 |
|
| 127 |
-
|
| 128 |
### Sea-Bench
|
| 129 |
|
| 130 |
|
|
|
|
| 47 |
- DEMO: [SeaLLMs/SeaLLM-7B](https://huggingface.co/spaces/SeaLLMs/SeaLLM-7B).
|
| 48 |
- Technical report: [Arxiv: SeaLLMs - Large Language Models for Southeast Asia](https://arxiv.org/pdf/2312.00738.pdf).
|
| 49 |
- Model weights:
|
| 50 |
+
- [SeaLLM-7B-v2.5](https://huggingface.co/SeaLLMs/SeaLLM-7B-v2.5).
|
| 51 |
+
- [SeaLLM-7B-v2.5-GGUF](https://huggingface.co/SeaLLMs/SeaLLM-7B-v2.5-GGUF).
|
| 52 |
- Run locally:
|
| 53 |
- [LM-studio](https://lmstudio.ai/):
|
| 54 |
- [SeaLLM-7B-v2.5-q4_0-chatml](https://huggingface.co/SeaLLMs/SeaLLM-7B-v2.5-GGUF/blob/main/seallm-7b-v2.5-chatml.Q4_K_M.gguf) with ChatML template (`<eos>` token changed to `<|im_end|>`)
|
| 55 |
- [SeaLLM-7B-v2.5-q4_0](https://huggingface.co/SeaLLMs/SeaLLM-7B-v2.5-GGUF/blob/main/seallm-7b-v2.5.Q4_K_M.gguf) - must use SeaLLM-7B-v2.5 chat format.
|
| 56 |
- [MLX for Apple Silicon](https://github.com/ml-explore/mlx): [SeaLLMs/SeaLLM-7B-v2.5-mlx-quantized](https://huggingface.co/SeaLLMs/SeaLLM-7B-v2.5-mlx-quantized)
|
| 57 |
+
- Previous models:
|
| 58 |
+
- [SeaLLM-7B-v2](https://huggingface.co/SeaLLMs/SeaLLM-7B-v2)
|
| 59 |
+
- [SeaLLM-7B-v1](https://huggingface.co/SeaLLMs/SeaLLM-7B-v1)
|
| 60 |
|
| 61 |
<blockquote style="color:red">
|
| 62 |
<p><strong style="color: red">Terms of Use and License</strong>:
|
|
|
|
| 112 |
| SeaLLM-7B-v2.5 | 78.5 | 34.9 | 51.3 | 22.1 | 72.3 | 30.2 | 71.5 | 30.1 | 62.0 | 28.4
|
| 113 |
|
| 114 |
|
|
|
|
| 115 |
Baselines were evaluated using their respective chat-template and system prompts ([Qwen1.5-7B-chat](https://huggingface.co/Qwen/Qwen1.5-7B-Chat/blob/main/tokenizer_config.json), [Vistral](https://huggingface.co/Viet-Mistral/Vistral-7B-Chat)).
|
| 116 |
|
| 117 |
#### Zero-shot MGSM
|
|
|
|
| 126 |
| SeaLLM-7B-v2.5 | 58.0 | **64.8**
|
| 127 |
|
| 128 |
|
|
|
|
| 129 |
### Sea-Bench
|
| 130 |
|
| 131 |
|