From c08e37f55d6d1e76cf8b2b79b78663c018409a65 Mon Sep 17 00:00:00 2001 From: localai-org-maint-bot <306269227+localai-org-maint-bot@users.noreply.github.com> Date: Sun, 25 Oct 2026 09:52:54 +0000 Subject: [PATCH] feat(gallery): add Sage-A1.5 GGUF variants Offer the official Q4_K_M and Q8_0 builds with their vision projector. Pin artifact revisions and document installation and variant selection. Assisted-by: Codex:gpt-6 (cherry picked from commit 9f98baa8e529583c2911d12539eabb33bca10225, mudler/LocalAI#12833) Assisted-by: Claude:claude-sonnet-5-5 Signed-off-by: Ettore Di Giacinto --- gallery/index.yaml | 82 ++++++++++++++++++++++++++++++++ gallery/models/index.yaml | 98 +++++++++++++++++++++++++++++++++++++++ 2 files changed, 180 insertions(+) diff --git a/gallery/index.yaml b/gallery/index.yaml index ac95dc7de..764c456ad 100644 --- a/gallery/index.yaml +++ b/gallery/index.yaml @@ -3,6 +3,88 @@ # written by scripts/models/cmd/modelcatalog and stays only so releases before # the cutover keep working. CI fails when it differs from the generated copy: # go run ./scripts/models/cmd/modelcatalog --library gallery/models/index.yaml --out gallery/index.yaml +- url: github:mudler/LocalAI/gallery/virtual.yaml@master + name: sage-a1.5-q4 + description: | + Sage-A1.5 is a 35B mixture-of-experts model based on Qwen3.6 for + scientific reasoning, coding, and tool use. This official Q4_K_M GGUF + build uses llama.cpp and includes the vision projector. + license: apache-2.0 + urls: + - https://huggingface.co/InternScience/Sage-A1.5 + - https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF + tags: + - llm + - gguf + - vision + - reasoning + - cpu + - gpu + files: + - filename: Sage-A1.5-Q4_K_M.gguf + sha256: a64fe0bbf936db3eb54f66f142c2b2eee569ade83673649f549a21daef8ca395 + uri: https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF/resolve/10b179b52f2c841030db327627e707d076b4c8cf/Sage-A1.5-Q4_K_M.gguf + - filename: Sage-A1.5-mmproj.gguf + sha256: 51cd4410057e8dc4152ca771a726fbfb34856f047e7049018c757d42ddd481f2 + uri: https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF/resolve/10b179b52f2c841030db327627e707d076b4c8cf/Sage-A1.5-mmproj.gguf + overrides: + backend: llama-cpp + context_size: 32768 + known_usecases: + - chat + - vision + mmproj: Sage-A1.5-mmproj.gguf + options: + - use_jinja:true + parameters: + model: Sage-A1.5-Q4_K_M.gguf + temperature: 1.0 + top_k: 20 + top_p: 0.95 + template: + use_tokenizer_template: true + variants: + - model: sage-a1.5-q8 +- url: github:mudler/LocalAI/gallery/virtual.yaml@master + name: sage-a1.5-q8 + description: | + Sage-A1.5 is a 35B mixture-of-experts model based on Qwen3.6 for + scientific reasoning, coding, and tool use. This official Q8_0 GGUF + build uses llama.cpp and includes the vision projector. + license: apache-2.0 + urls: + - https://huggingface.co/InternScience/Sage-A1.5 + - https://huggingface.co/InternScience/Sage-A1.5-Q8_0-GGUF + tags: + - llm + - gguf + - vision + - reasoning + - cpu + - gpu + files: + - filename: Sage-A1.5-Q8_0.gguf + sha256: f7247f6faef3f6c616b953595e1fc7c3fbf257d02d9915af5fdf15e808ea8ee4 + uri: https://huggingface.co/InternScience/Sage-A1.5-Q8_0-GGUF/resolve/1cbe2925983a13a6c09881b9bc088884dbfbc146/Sage-A1.5-Q8_0.gguf + - filename: Sage-A1.5-mmproj.gguf + sha256: 51cd4410057e8dc4152ca771a726fbfb34856f047e7049018c757d42ddd481f2 + uri: https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF/resolve/10b179b52f2c841030db327627e707d076b4c8cf/Sage-A1.5-mmproj.gguf + overrides: + backend: llama-cpp + context_size: 32768 + known_usecases: + - chat + - vision + mmproj: Sage-A1.5-mmproj.gguf + options: + - use_jinja:true + parameters: + model: Sage-A1.5-Q8_0.gguf + temperature: 1.0 + top_k: 20 + top_p: 0.95 + template: + use_tokenizer_template: true - url: github:mudler/LocalAI/gallery/virtual.yaml@master name: lfm2.5-24b-a2b-turbo-brilliance-iq4-xs description: | diff --git a/gallery/models/index.yaml b/gallery/models/index.yaml index 458fd9bd5..9410760d6 100644 --- a/gallery/models/index.yaml +++ b/gallery/models/index.yaml @@ -4,6 +4,104 @@ # changes and never edit it by hand. CI fails when it is stale. A build from # master reads it as the default model source; a tagged build reads the gallery # until the release cutover. +- kind: model + id: sage-a1.5-q4 + name: sage-a1.5-q4 + version: 0.0.0-gallery.88087a1aab17 + description: | + Sage-A1.5 is a 35B mixture-of-experts model based on Qwen3.6 for + scientific reasoning, coding, and tool use. This official Q4_K_M GGUF + build uses llama.cpp and includes the vision projector. + license: apache-2.0 + urls: + - https://huggingface.co/InternScience/Sage-A1.5 + - https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF + tags: + - llm + - gguf + - vision + - reasoning + - cpu + - gpu + channel: gallery + policy: + requireSignature: false + schemaVersion: 1 + model: + url: github:mudler/LocalAI/gallery/virtual.yaml@master + files: + - filename: Sage-A1.5-Q4_K_M.gguf + sha256: a64fe0bbf936db3eb54f66f142c2b2eee569ade83673649f549a21daef8ca395 + uri: https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF/resolve/10b179b52f2c841030db327627e707d076b4c8cf/Sage-A1.5-Q4_K_M.gguf + - filename: Sage-A1.5-mmproj.gguf + sha256: 51cd4410057e8dc4152ca771a726fbfb34856f047e7049018c757d42ddd481f2 + uri: https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF/resolve/10b179b52f2c841030db327627e707d076b4c8cf/Sage-A1.5-mmproj.gguf + overrides: + backend: llama-cpp + context_size: 32768 + known_usecases: + - chat + - vision + mmproj: Sage-A1.5-mmproj.gguf + options: + - use_jinja:true + parameters: + model: Sage-A1.5-Q4_K_M.gguf + temperature: 1.0 + top_k: 20 + top_p: 0.95 + template: + use_tokenizer_template: true + variants: + - model: sage-a1.5-q8 +- kind: model + id: sage-a1.5-q8 + name: sage-a1.5-q8 + version: 0.0.0-gallery.d2ee65a3043e + description: | + Sage-A1.5 is a 35B mixture-of-experts model based on Qwen3.6 for + scientific reasoning, coding, and tool use. This official Q8_0 GGUF + build uses llama.cpp and includes the vision projector. + license: apache-2.0 + urls: + - https://huggingface.co/InternScience/Sage-A1.5 + - https://huggingface.co/InternScience/Sage-A1.5-Q8_0-GGUF + tags: + - llm + - gguf + - vision + - reasoning + - cpu + - gpu + channel: gallery + policy: + requireSignature: false + schemaVersion: 1 + model: + url: github:mudler/LocalAI/gallery/virtual.yaml@master + files: + - filename: Sage-A1.5-Q8_0.gguf + sha256: f7247f6faef3f6c616b953595e1fc7c3fbf257d02d9915af5fdf15e808ea8ee4 + uri: https://huggingface.co/InternScience/Sage-A1.5-Q8_0-GGUF/resolve/1cbe2925983a13a6c09881b9bc088884dbfbc146/Sage-A1.5-Q8_0.gguf + - filename: Sage-A1.5-mmproj.gguf + sha256: 51cd4410057e8dc4152ca771a726fbfb34856f047e7049018c757d42ddd481f2 + uri: https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF/resolve/10b179b52f2c841030db327627e707d076b4c8cf/Sage-A1.5-mmproj.gguf + overrides: + backend: llama-cpp + context_size: 32768 + known_usecases: + - chat + - vision + mmproj: Sage-A1.5-mmproj.gguf + options: + - use_jinja:true + parameters: + model: Sage-A1.5-Q8_0.gguf + temperature: 1.0 + top_k: 20 + top_p: 0.95 + template: + use_tokenizer_template: true - kind: model id: lfm2.5-24b-a2b-turbo-brilliance-iq4-xs name: lfm2.5-24b-a2b-turbo-brilliance-iq4-xs