diff --git a/.github/workflows/codegen.yaml b/.github/workflows/codegen.yaml index 5a3a5cd8..ce59f556 100644 --- a/.github/workflows/codegen.yaml +++ b/.github/workflows/codegen.yaml @@ -63,7 +63,7 @@ jobs: - name: Download model working-directory: ./models - run: make download-model-mistral + run: make download-model-mistral-code - name: Run Functional Tests shell: bash diff --git a/.github/workflows/model_image_build_push.yaml b/.github/workflows/model_image_build_push.yaml index 5eaa73f0..685b504b 100644 --- a/.github/workflows/model_image_build_push.yaml +++ b/.github/workflows/model_image_build_push.yaml @@ -43,6 +43,9 @@ jobs: label: Q4_K_M url: https://huggingface.co/instructlab/granite-7b-lab-GGUF/resolve/main/granite-7b-lab-Q4_K_M.gguf platforms: linux/amd64,linux/arm64 + - image_name: whisper-small + url: https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-small.bin + platforms: linux/amd64,linux/arm64 runs-on: ubuntu-latest permissions: contents: read diff --git a/models/Containerfile b/models/Containerfile index b53ab9c2..e00a0edd 100644 --- a/models/Containerfile +++ b/models/Containerfile @@ -1,8 +1,8 @@ # Suggested alternative open AI Models -# https://huggingface.co/instructlab/granite-7b-lab-GGUF/resolve/main/granite-7b-lab-Q4_K_M.gguf +# https://huggingface.co/instructlab/granite-7b-lab-GGUF/resolve/main/granite-7b-lab-Q4_K_M.gguf (Default) # https://huggingface.co/instructlab/merlinite-7b-lab-GGUF/resolve/main/merlinite-7b-lab-Q4_K_M.gguf # https://huggingface.co/TheBloke/Llama-2-7B-Chat-GGUF/resolve/main/llama-2-7b-chat.Q5_K_S.gguf -# https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/resolve/main/mistral-7b-instruct-v0.2.Q4_K_M.gguf (Default) +# https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/resolve/main/mistral-7b-instruct-v0.2.Q4_K_M.gguf # https://huggingface.co/TheBloke/CodeLlama-7B-Instruct-GGUF/resolve/main/codellama-7b-instruct.Q4_K_M.gguf # https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-small.bin # podman build --build-arg="MODEL_URL=https://..." -t quay.io/yourimage . diff --git a/models/Makefile b/models/Makefile index e58eb647..76f2e669 100644 --- a/models/Makefile +++ b/models/Makefile @@ -25,7 +25,7 @@ download-model-granite: download-model-merlinite: $(MAKE) MODEL_URL=https://huggingface.co/instructlab/merlinite-7b-lab-GGUF/resolve/main/merlinite-7b-lab-Q4_K_M.gguf MODEL_NAME=merlinite-7b-lab-Q4_K_M.gguf download-model -.PHONY: download-model-whisper-small # small .bin model type testing +.PHONY: download-model-whisper-small download-model-whisper-small: $(MAKE) MODEL_NAME=ggml-small.bin MODEL_URL=https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-small.bin download-model @@ -33,7 +33,11 @@ download-model-whisper-small: download-model-mistral: $(MAKE) MODEL_NAME=mistral-7b-instruct-v0.2.Q4_K_M.gguf MODEL_URL=https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/resolve/main/mistral-7b-instruct-v0.2.Q4_K_M.gguf download-model +.PHONY: download-model-mistral-code +download-mode-mistral-code: + $(MAKE) MODEL_NAME=mistral-7b-code-16k-qlora.Q4_K_M.gguf MODEL_URL=https://huggingface.co/TheBloke/Mistral-7B-Code-16K-qlora-GGUF/resolve/main/mistral-7b-code-16k-qlora.Q4_K_M.gguf + .PHONY: clean clean: - rm -f *tmp - rm -f mistral* whisper* granite* merlinite* + -rm -f *tmp + -rm -f mistral* ggml-* granite* merlinite* diff --git a/models/README.md b/models/README.md index c8f6c27f..f2ecaf9f 100644 --- a/models/README.md +++ b/models/README.md @@ -8,6 +8,7 @@ Want to try one of our tested models? Try one or all of the following: make download-model-granite make download-model-merlinite make download-model-mistral +make download-model-mistral-code make download-model-whisper-small ``` diff --git a/recipes/audio/audio_to_text/bootc/Containerfile b/recipes/audio/audio_to_text/bootc/Containerfile index d53c26d1..6a6c0921 100644 --- a/recipes/audio/audio_to_text/bootc/Containerfile +++ b/recipes/audio/audio_to_text/bootc/Containerfile @@ -14,9 +14,9 @@ RUN set -eu; mkdir -p /usr/ssh && \ echo ${SSHPUBKEY} > /usr/ssh/root.keys && chmod 0600 /usr/ssh/root.keys ARG RECIPE=audio-to-text -ARG MODEL_IMAGE=quay.io/ai-lab/mistral-7b-instruct:latest +ARG MODEL_IMAGE=quay.io/ai-lab/whisper-small:latest ARG APP_IMAGE=quay.io/ai-lab/${RECIPE}:latest -ARG SERVER_IMAGE=quay.io/ai-lab/llamacpp_python:latest +ARG SERVER_IMAGE=quay.io/ai-lab/whispercpp:latest ARG TARGETARCH # Add quadlet files to setup system to automatically run AI application on boot