From de09ae42ef62fd4156c7220e9e3f11ac88988c59 Mon Sep 17 00:00:00 2001 From: "LocalAI [bot]" <139863280+localai-bot@users.noreply.github.com> Date: Tue, 18 Feb 2025 09:11:07 +0100 Subject: [PATCH 1/7] chore: :arrow_up: Update ggml-org/llama.cpp to `73e2ed3ce3492d3ed70193dd09ae8aa44779651d` (#4854) :arrow_up: Update ggml-org/llama.cpp Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 7e4ba248..e0516b2d 100644 --- a/Makefile +++ b/Makefile @@ -6,7 +6,7 @@ BINARY_NAME=local-ai DETECT_LIBS?=true # llama.cpp versions -CPPLLAMA_VERSION?=2eea03d86a2d132c8245468c26290ce07a27a8e8 +CPPLLAMA_VERSION?=73e2ed3ce3492d3ed70193dd09ae8aa44779651d # whisper.cpp version WHISPER_REPO?=https://github.com/ggerganov/whisper.cpp From 7e76ea40fbac8049bb46cd6ff0cefe8eaf0f4fd5 Mon Sep 17 00:00:00 2001 From: Ettore Di Giacinto Date: Tue, 18 Feb 2025 11:47:00 +0100 Subject: [PATCH 2/7] chore(model gallery): add kubeguru-llama3.2-3b-v0.1 (#4858) Signed-off-by: Ettore Di Giacinto --- gallery/index.yaml | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/gallery/index.yaml b/gallery/index.yaml index b8d6a66f..f4661241 100644 --- a/gallery/index.yaml +++ b/gallery/index.yaml @@ -1550,6 +1550,23 @@ - filename: localai-functioncall-llama3.2-3b-v0.5-q4_k_m.gguf sha256: edc50f6c243e6bd6912599661a15e030de03d2be53409663ac27d3ca48306ee4 uri: huggingface://mudler/LocalAI-functioncall-llama3.2-3b-v0.5-Q4_K_M-GGUF/localai-functioncall-llama3.2-3b-v0.5-q4_k_m.gguf +- !!merge <<: *llama32 + name: "kubeguru-llama3.2-3b-v0.1" + icon: https://cdn-uploads.huggingface.co/production/uploads/647374aa7ff32a81ac6d35d4/rptpRyhrcUEG3i2OPT897.png + urls: + - https://huggingface.co/Spectro-Cloud/kubeguru-llama3.2-3b-v0.1 + - https://huggingface.co/mradermacher/kubeguru-llama3.2-3b-v0.1-GGUF + description: | + Kubeguru: Your Kubernetes & Linux Expert AI + Ask anything about Kubernetes, Linux, containers—and get expert answers in real-time! + Kubeguru is a specialized Large Language Model (LLM) developed and released by the Open Source team at Spectro Cloud. Whether you're managing cloud-native applications, deploying edge workloads, or troubleshooting containerized services, Kubeguru provides precise, actionable insights at every step. + overrides: + parameters: + model: kubeguru-llama3.2-3b-v0.1.Q4_K_M.gguf + files: + - filename: kubeguru-llama3.2-3b-v0.1.Q4_K_M.gguf + sha256: 770900ba9594f64f31b35fe444d31263712cabe167efaf4201d79fdc29de9533 + uri: huggingface://mradermacher/kubeguru-llama3.2-3b-v0.1-GGUF/kubeguru-llama3.2-3b-v0.1.Q4_K_M.gguf - &qwen25 name: "qwen2.5-14b-instruct" ## Qwen2.5 icon: https://avatars.githubusercontent.com/u/141221163 From fac70e9642b6dab31fa5d8ba4bab9a20b1031316 Mon Sep 17 00:00:00 2001 From: Ettore Di Giacinto Date: Tue, 18 Feb 2025 11:49:26 +0100 Subject: [PATCH 3/7] chore(model gallery): add allenai_llama-3.1-tulu-3.1-8b (#4859) Signed-off-by: Ettore Di Giacinto --- gallery/index.yaml | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/gallery/index.yaml b/gallery/index.yaml index f4661241..beb267cd 100644 --- a/gallery/index.yaml +++ b/gallery/index.yaml @@ -5828,6 +5828,23 @@ - filename: DavidBrowne17_LlamaThink-8B-instruct-Q4_K_M.gguf sha256: 6aea4e13f03347e03d6989c736a7ccab82582115eb072cacfeb7f0b645a8bec0 uri: huggingface://bartowski/DavidBrowne17_LlamaThink-8B-instruct-GGUF/DavidBrowne17_LlamaThink-8B-instruct-Q4_K_M.gguf +- !!merge <<: *llama31 + name: "allenai_llama-3.1-tulu-3.1-8b" + icon: https://huggingface.co/datasets/allenai/blog-images/resolve/main/tulu3/Tulu3-logo.png + urls: + - https://huggingface.co/allenai/Llama-3.1-Tulu-3.1-8B + - https://huggingface.co/bartowski/allenai_Llama-3.1-Tulu-3.1-8B-GGUF + description: | + Tülu 3 is a leading instruction following model family, offering a post-training package with fully open-source data, code, and recipes designed to serve as a comprehensive guide for modern techniques. This is one step of a bigger process to training fully open-source models, like our OLMo models. Tülu 3 is designed for state-of-the-art performance on a diversity of tasks in addition to chat, such as MATH, GSM8K, and IFEval. + + Version 3.1 update: The new version of our Tülu model is from an improvement only in the final RL stage of training. We switched from PPO to GRPO (no reward model) and did further hyperparameter tuning to achieve substantial performance improvements across the board over the original Tülu 3 8B model. + overrides: + parameters: + model: allenai_Llama-3.1-Tulu-3.1-8B-Q4_K_M.gguf + files: + - filename: allenai_Llama-3.1-Tulu-3.1-8B-Q4_K_M.gguf + sha256: 5eae0f1a9bcdea7cad9f1d0d5ba7540bb3de3e2d72293c076a23f24db1c2c7da + uri: huggingface://bartowski/allenai_Llama-3.1-Tulu-3.1-8B-GGUF/allenai_Llama-3.1-Tulu-3.1-8B-Q4_K_M.gguf - &deepseek url: "github:mudler/LocalAI/gallery/deepseek.yaml@master" ## Deepseek name: "deepseek-coder-v2-lite-instruct" From 131ea5b6271d89b13b34e5c9670e7f5b2aaf47c7 Mon Sep 17 00:00:00 2001 From: Ettore Di Giacinto Date: Tue, 18 Feb 2025 11:51:29 +0100 Subject: [PATCH 4/7] chore(model gallery): add nbeerbower_dumpling-qwen2.5-14b (#4860) Signed-off-by: Ettore Di Giacinto --- gallery/index.yaml | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/gallery/index.yaml b/gallery/index.yaml index beb267cd..57416655 100644 --- a/gallery/index.yaml +++ b/gallery/index.yaml @@ -3847,6 +3847,34 @@ - filename: ozone-ai_0x-lite-Q4_K_M.gguf sha256: 7f163e72ead7522bd6774555a932e0a11f212d17cdc9442e2cfd1b017009f832 uri: huggingface://bartowski/ozone-ai_0x-lite-GGUF/ozone-ai_0x-lite-Q4_K_M.gguf +- !!merge <<: *qwen25 + name: "nbeerbower_dumpling-qwen2.5-14b" + icon: https://huggingface.co/nbeerbower/Dumpling-Qwen2.5-32B/resolve/main/dumpling_cover.png?download=true + urls: + - https://huggingface.co/nbeerbower/Dumpling-Qwen2.5-14B + - https://huggingface.co/bartowski/nbeerbower_Dumpling-Qwen2.5-14B-GGUF + description: | + nbeerbower/EVA-abliterated-TIES-Qwen2.5-14B finetuned on: + + nbeerbower/GreatFirewall-DPO + nbeerbower/Schule-DPO + nbeerbower/Purpura-DPO + nbeerbower/Arkhaios-DPO + jondurbin/truthy-dpo-v0.1 + antiven0m/physical-reasoning-dpo + flammenai/Date-DPO-NoAsterisks + flammenai/Prude-Phi3-DPO + Atsunori/HelpSteer2-DPO + jondurbin/gutenberg-dpo-v0.1 + nbeerbower/gutenberg2-dpo + nbeerbower/gutenberg-moderne-dpo. + overrides: + parameters: + model: nbeerbower_Dumpling-Qwen2.5-14B-Q4_K_M.gguf + files: + - filename: nbeerbower_Dumpling-Qwen2.5-14B-Q4_K_M.gguf + sha256: 2d38348414b2719971a08a604313ed98b44b586490633d6e237dd096ae5bf31d + uri: huggingface://bartowski/nbeerbower_Dumpling-Qwen2.5-14B-GGUF/nbeerbower_Dumpling-Qwen2.5-14B-Q4_K_M.gguf - &llama31 url: "github:mudler/LocalAI/gallery/llama3.1-instruct.yaml@master" ## LLama3.1 icon: https://avatars.githubusercontent.com/u/153379578 From b9b5a635ca2c578006e6a2eb1b7093f863da619f Mon Sep 17 00:00:00 2001 From: Ettore Di Giacinto Date: Tue, 18 Feb 2025 11:53:23 +0100 Subject: [PATCH 5/7] chore(model gallery): add nbeerbower_dumpling-qwen2.5-32b-v2 (#4861) Signed-off-by: Ettore Di Giacinto --- gallery/index.yaml | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/gallery/index.yaml b/gallery/index.yaml index 57416655..7f43c2ba 100644 --- a/gallery/index.yaml +++ b/gallery/index.yaml @@ -3875,6 +3875,34 @@ - filename: nbeerbower_Dumpling-Qwen2.5-14B-Q4_K_M.gguf sha256: 2d38348414b2719971a08a604313ed98b44b586490633d6e237dd096ae5bf31d uri: huggingface://bartowski/nbeerbower_Dumpling-Qwen2.5-14B-GGUF/nbeerbower_Dumpling-Qwen2.5-14B-Q4_K_M.gguf +- !!merge <<: *qwen25 + name: "nbeerbower_dumpling-qwen2.5-32b-v2" + icon: https://huggingface.co/nbeerbower/Dumpling-Qwen2.5-32B/resolve/main/dumpling_cover.png?download=true + urls: + - https://huggingface.co/nbeerbower/Dumpling-Qwen2.5-32B-v2 + - https://huggingface.co/bartowski/nbeerbower_Dumpling-Qwen2.5-32B-v2-GGUF + description: | + nbeerbower/Rombos-EVAGutenberg-TIES-Qwen2.5-32B finetuned on: + + nbeerbower/GreatFirewall-DPO + nbeerbower/Schule-DPO + nbeerbower/Purpura-DPO + nbeerbower/Arkhaios-DPO + jondurbin/truthy-dpo-v0.1 + antiven0m/physical-reasoning-dpo + flammenai/Date-DPO-NoAsterisks + flammenai/Prude-Phi3-DPO + Atsunori/HelpSteer2-DPO + jondurbin/gutenberg-dpo-v0.1 + nbeerbower/gutenberg2-dpo + nbeerbower/gutenberg-moderne-dpo. + overrides: + parameters: + model: nbeerbower_Dumpling-Qwen2.5-32B-v2-Q4_K_M.gguf + files: + - filename: nbeerbower_Dumpling-Qwen2.5-32B-v2-Q4_K_M.gguf + sha256: 02a5320d62e13b31ac6d04ccdaba7b72a524d6cc72a7082b94d8cac0a183ecb4 + uri: huggingface://bartowski/nbeerbower_Dumpling-Qwen2.5-32B-v2-GGUF/nbeerbower_Dumpling-Qwen2.5-32B-v2-Q4_K_M.gguf - &llama31 url: "github:mudler/LocalAI/gallery/llama3.1-instruct.yaml@master" ## LLama3.1 icon: https://avatars.githubusercontent.com/u/153379578 From bb85b6ef00656ba29019cb78c1fb5f5a7814335b Mon Sep 17 00:00:00 2001 From: Ettore Di Giacinto Date: Tue, 18 Feb 2025 12:44:44 +0100 Subject: [PATCH 6/7] feat: improve ui models list in the index (#4863) * feat(ui): improve index - Redirect to the chat view when clicking on a model Signed-off-by: Ettore Di Giacinto * Display chat icon nearby the model Signed-off-by: Ettore Di Giacinto --------- Signed-off-by: Ettore Di Giacinto --- core/http/views/chat.html | 2 +- core/http/views/index.html | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/core/http/views/chat.html b/core/http/views/chat.html index b0f11281..627aff8a 100644 --- a/core/http/views/chat.html +++ b/core/http/views/chat.html @@ -44,7 +44,7 @@ SOFTWARE.
-

Chat with {{.Model}} +

Chat with {{.Model}}