From 14e8b040e9f52b4d5182e421d34107ab769f66ef Mon Sep 17 00:00:00 2001 From: Aaron Pham Date: Thu, 26 Sep 2024 02:32:45 -0400 Subject: [PATCH] chore(readme): remove older models from feed (#22) Signed-off-by: Aaron Pham --- README.md | 74 +++++++++++++++++++++++++-------------------------- gen_readme.py | 19 +++++-------- 2 files changed, 44 insertions(+), 49 deletions(-) diff --git a/README.md b/README.md index d1a0305c..77e30fef 100644 --- a/README.md +++ b/README.md @@ -16,17 +16,17 @@ openllm repo add nightly https://github.com/bentoml/openllm-models@nightly - [Llama-3.2](#llama3.2) - [Llama-3.1](#llama3.1) -- [Llama-3](#llama3) - [Phi-3](#phi3) - [Mistral](#mistral) - [Gemma-2](#gemma2) - [Qwen-2.5](#qwen2.5) -- [Qwen-2](#qwen2) -- [Gemma](#gemma) -- [Llama-2](#llama2) - [Mixtral](#mixtral) - [Mistral-Large](#mistral-large) - [Codestral](#codestral) +- [](#llama2) +- [Gemma](#gemma) +- [Llama-3](#llama3) +- [Qwen-2](#qwen2) --- @@ -54,18 +54,6 @@ openllm repo add nightly https://github.com/bentoml/openllm-models@nightly --- -### Llama-3 - -| Model | Version | Huggingface Link | -| --- | --- | --- | -| llama3 | 70b-instruct-awq-4bit-e96c | [HF Link](https://huggingface.co/casperhansen/llama-3-70b-instruct-awq) | -| llama3 | 70b-instruct-fp16-45fe | [HF Link](https://huggingface.co/meta-llama/Meta-Llama-3-70B-Instruct) | -| llama3 | 8b-instruct-awq-4bit-b159 | [HF Link](https://huggingface.co/casperhansen/llama-3-8b-instruct-awq) | -| llama3 | 8b-instruct-fp16-72f8 | [HF Link](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) | - ---- - - ### Phi-3 | Model | Version | Huggingface Link | @@ -112,33 +100,36 @@ openllm repo add nightly https://github.com/bentoml/openllm-models@nightly --- -### Qwen-2 +### Mixtral | Model | Version | Huggingface Link | | --- | --- | --- | -| qwen2 | 0.5b-instruct-fp16-114e | [HF Link](https://huggingface.co/Qwen/Qwen2-0.5B-Instruct) | -| qwen2 | 1.5b-instruct-fp16-743d | [HF Link](https://huggingface.co/Qwen/Qwen2-1.5B-Instruct) | -| qwen2 | 57b-a14b-instruct-fp16-13ca | [HF Link](https://huggingface.co/Qwen/Qwen2-57B-A14B-Instruct) | -| qwen2 | 72b-instruct-awq-4bit-5384 | [HF Link](https://huggingface.co/Qwen/Qwen2-72B-Instruct-AWQ) | -| qwen2 | 72b-instruct-fp16-4755 | [HF Link](https://huggingface.co/Qwen/Qwen2-72B-Instruct) | -| qwen2 | 7b-instruct-awq-4bit-89de | [HF Link](https://huggingface.co/Qwen/Qwen2-7B-Instruct-AWQ) | -| qwen2 | 7b-instruct-fp16-c17f | [HF Link](https://huggingface.co/Qwen/Qwen2-7B-Instruct) | +| mixtral | 8x7b-instruct-v0.1-awq-4bit-2117 | [HF Link](https://huggingface.co/casperhansen/mixtral-instruct-awq) | +| mixtral | 8x7b-instruct-v0.1-fp16-55c3 | [HF Link](https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1) | --- -### Gemma +### Mistral-Large | Model | Version | Huggingface Link | | --- | --- | --- | -| gemma | 2b-instruct-fp16-e12f | [HF Link](https://huggingface.co/google/gemma-2b-it) | -| gemma | 7b-instruct-awq-4bit-1134 | [HF Link](https://huggingface.co/casperhansen/gemma-7b-it-awq) | -| gemma | 7b-instruct-fp16-e12a | [HF Link](https://huggingface.co/google/gemma-7b-it) | +| mistral-large | 123b-instruct-awq-4bit-e339 | [HF Link](https://huggingface.co/casperhansen/mistral-large-instruct-2407-awq) | +| mistral-large | 123b-instruct-fp16-eb4a | [HF Link](https://huggingface.co/mistralai/Mistral-Large-Instruct-2407) | + +--- + + +### Codestral + +| Model | Version | Huggingface Link | +| --- | --- | --- | +| codestral | 22b-v0.1-fp16-0d5b | [HF Link](https://huggingface.co/mistralai/Codestral-22B-v0.1) | --- -### Llama-2 +### | Model | Version | Huggingface Link | | --- | --- | --- | @@ -150,31 +141,40 @@ openllm repo add nightly https://github.com/bentoml/openllm-models@nightly --- -### Mixtral +### Gemma | Model | Version | Huggingface Link | | --- | --- | --- | -| mixtral | 8x7b-instruct-v0.1-awq-4bit-2117 | [HF Link](https://huggingface.co/casperhansen/mixtral-instruct-awq) | -| mixtral | 8x7b-instruct-v0.1-fp16-55c3 | [HF Link](https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1) | +| gemma | 2b-instruct-fp16-e12f | [HF Link](https://huggingface.co/google/gemma-2b-it) | +| gemma | 7b-instruct-awq-4bit-1134 | [HF Link](https://huggingface.co/casperhansen/gemma-7b-it-awq) | +| gemma | 7b-instruct-fp16-e12a | [HF Link](https://huggingface.co/google/gemma-7b-it) | --- -### Mistral-Large +### Llama-3 | Model | Version | Huggingface Link | | --- | --- | --- | -| mistral-large | 123b-instruct-awq-4bit-e339 | [HF Link](https://huggingface.co/casperhansen/mistral-large-instruct-2407-awq) | -| mistral-large | 123b-instruct-fp16-eb4a | [HF Link](https://huggingface.co/mistralai/Mistral-Large-Instruct-2407) | +| llama3 | 70b-instruct-awq-4bit-e96c | [HF Link](https://huggingface.co/casperhansen/llama-3-70b-instruct-awq) | +| llama3 | 70b-instruct-fp16-45fe | [HF Link](https://huggingface.co/meta-llama/Meta-Llama-3-70B-Instruct) | +| llama3 | 8b-instruct-awq-4bit-b159 | [HF Link](https://huggingface.co/casperhansen/llama-3-8b-instruct-awq) | +| llama3 | 8b-instruct-fp16-72f8 | [HF Link](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) | --- -### Codestral +### Qwen-2 | Model | Version | Huggingface Link | | --- | --- | --- | -| codestral | 22b-v0.1-fp16-0d5b | [HF Link](https://huggingface.co/mistralai/Codestral-22B-v0.1) | +| qwen2 | 0.5b-instruct-fp16-114e | [HF Link](https://huggingface.co/Qwen/Qwen2-0.5B-Instruct) | +| qwen2 | 1.5b-instruct-fp16-743d | [HF Link](https://huggingface.co/Qwen/Qwen2-1.5B-Instruct) | +| qwen2 | 57b-a14b-instruct-fp16-13ca | [HF Link](https://huggingface.co/Qwen/Qwen2-57B-A14B-Instruct) | +| qwen2 | 72b-instruct-awq-4bit-5384 | [HF Link](https://huggingface.co/Qwen/Qwen2-72B-Instruct-AWQ) | +| qwen2 | 72b-instruct-fp16-4755 | [HF Link](https://huggingface.co/Qwen/Qwen2-72B-Instruct) | +| qwen2 | 7b-instruct-awq-4bit-89de | [HF Link](https://huggingface.co/Qwen/Qwen2-7B-Instruct-AWQ) | +| qwen2 | 7b-instruct-fp16-c17f | [HF Link](https://huggingface.co/Qwen/Qwen2-7B-Instruct) | --- diff --git a/gen_readme.py b/gen_readme.py index 64af4075..8cd67e25 100644 --- a/gen_readme.py +++ b/gen_readme.py @@ -17,7 +17,6 @@ "qwen2.5": "Qwen-2.5", "gemma": "Gemma", "gemma2": "Gemma-2", - "llama2": "Llama-2", "mixtral": "Mixtral", "mistral-large": "Mistral-Large", "codestral": "Codestral", @@ -26,17 +25,13 @@ model_priority = { "llama3.2": 1, "llama3.1": 2, - "llama3": 3, - "phi3": 4, - "mistral": 5, - "gemma2": 6, - "qwen2.5": 7, - "qwen2": 8, - "gemma": 9, - "llama2": 10, - "mixtral": 11, - "mistral-large": 12, - "codestral": 13, + "phi3": 3, + "mistral": 4, + "gemma2": 5, + "qwen2.5": 6, + "mixtral": 7, + "mistral-large": 8, + "codestral": 9, } yaml_files = glob.glob(os.path.join(root_dir, "**/bento.yaml"), recursive=True)