From 7270bdc508ea76a0779c9a3c576aa21df6b8757a Mon Sep 17 00:00:00 2001 From: localai-org-maint-bot <306269227+localai-org-maint-bot@users.noreply.github.com> Date: Mon, 19 Oct 2026 16:05:02 +0000 Subject: [PATCH] feat(gallery): add Ternary Bonsai 2 2B variants Add PQ2_0, PTQ1_0, and Q2_0 builds with the Q8_0 vision projector. Pin files to the publisher revision and document variant selection. Assisted-by: Codex:gpt-6 --- docs/content/features/model-gallery.md | 18 +++ gallery/index.yaml | 144 +++++++++++++++++++++ gallery/models/index.yaml | 168 +++++++++++++++++++++++++ 3 files changed, 330 insertions(+) diff --git a/docs/content/features/model-gallery.md b/docs/content/features/model-gallery.md index 56dde2118..68b3af0be 100644 --- a/docs/content/features/model-gallery.md +++ b/docs/content/features/model-gallery.md @@ -44,6 +44,24 @@ Both views store the view, search, filter, and selection in the URL. Installing from Explore does not move you away from the catalog; the entry updates in place when the operation finishes. +### Ternary Bonsai 2 2B + +[Ternary Bonsai 2 2B](https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf) +is a vision-language model derived from Qwen3.5-2B. The gallery offers PQ2_0, +PTQ1_0, and Q2_0 packings through the Bonsai backend. Each includes the Q8_0 +vision projector and uses a 32,768-token default context. These packings contain +the same ternary weights in different storage layouts. + +Install with automatic variant selection: + +```bash +local-ai models install ternary-bonsai-2-2b +``` + +Use `--variant ternary-bonsai-2-2b-ptq1` for the smallest packing or +`--variant ternary-bonsai-2-2b-q2` for Q2_0. To select PQ2_0 explicitly, +use `--variant ternary-bonsai-2-2b`. + ### MiniCPM-V 4.7 1B [MiniCPM-V 4.7 1B](https://huggingface.co/openbmb/MiniCPM-V-4.7-1B) diff --git a/gallery/index.yaml b/gallery/index.yaml index 981cc3563..b976ec953 100644 --- a/gallery/index.yaml +++ b/gallery/index.yaml @@ -3,6 +3,150 @@ # 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: ternary-bonsai-2-2b + description: | + Ternary Bonsai 2 2B is a vision-language model derived from Qwen3.5-2B + with ternary transformer weights. This PQ2_0 build uses the Bonsai + backend and includes the Q8_0 vision projector. The three standard + packings contain the same weights in different storage layouts. + license: apache-2.0 + urls: + - https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf + - https://huggingface.co/Qwen/Qwen3.5-2B + tags: + - llm + - gguf + - reasoning + - vision + - multimodal + files: + - filename: bonsai/models/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-PQ2_0.gguf + sha256: d5f01a4ecaf49239653308ee732ad1a0c5f802c0ddb997549fc7b694c8b45928 + uri: https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf/resolve/683f7be2629342a50987627b40881e54761a2430/Ternary-Bonsai-2-2B-PQ2_0.gguf + - filename: bonsai/mmproj/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + sha256: ac9934e8d7966d0d62442c202ec8fdad99a045dea79366c1a4a246a46f5c0533 + uri: https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf/resolve/683f7be2629342a50987627b40881e54761a2430/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + overrides: + backend: bonsai + context_size: 32768 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: bonsai/mmproj/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + options: + - use_jinja:true + parameters: + min_p: 0.0 + model: bonsai/models/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-PQ2_0.gguf + presence_penalty: 1.0 + repeat_penalty: 1.0 + temperature: 1.0 + top_k: 20 + top_p: 0.95 + template: + use_tokenizer_template: true + variants: + - model: ternary-bonsai-2-2b-ptq1 + - model: ternary-bonsai-2-2b-q2 +- url: github:mudler/LocalAI/gallery/virtual.yaml@master + name: ternary-bonsai-2-2b-ptq1 + description: | + Ternary Bonsai 2 2B is a vision-language model derived from Qwen3.5-2B + with ternary transformer weights. This PTQ1_0 build uses the Bonsai + backend and includes the Q8_0 vision projector. The three standard + packings contain the same weights in different storage layouts. + license: apache-2.0 + urls: + - https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf + - https://huggingface.co/Qwen/Qwen3.5-2B + tags: + - llm + - gguf + - reasoning + - vision + - multimodal + files: + - filename: bonsai/models/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-PTQ1_0.gguf + sha256: 36061ad406df8a4903a6db005be350c50fabfaf717f9ef0acf4b76adcd9929fa + uri: https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf/resolve/683f7be2629342a50987627b40881e54761a2430/Ternary-Bonsai-2-2B-PTQ1_0.gguf + - filename: bonsai/mmproj/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + sha256: ac9934e8d7966d0d62442c202ec8fdad99a045dea79366c1a4a246a46f5c0533 + uri: https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf/resolve/683f7be2629342a50987627b40881e54761a2430/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + overrides: + backend: bonsai + context_size: 32768 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: bonsai/mmproj/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + options: + - use_jinja:true + parameters: + min_p: 0.0 + model: bonsai/models/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-PTQ1_0.gguf + presence_penalty: 1.0 + repeat_penalty: 1.0 + temperature: 1.0 + top_k: 20 + top_p: 0.95 + template: + use_tokenizer_template: true +- url: github:mudler/LocalAI/gallery/virtual.yaml@master + name: ternary-bonsai-2-2b-q2 + description: | + Ternary Bonsai 2 2B is a vision-language model derived from Qwen3.5-2B + with ternary transformer weights. This Q2_0 build uses the Bonsai + backend and includes the Q8_0 vision projector. The three standard + packings contain the same weights in different storage layouts. + license: apache-2.0 + urls: + - https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf + - https://huggingface.co/Qwen/Qwen3.5-2B + tags: + - llm + - gguf + - reasoning + - vision + - multimodal + files: + - filename: bonsai/models/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-Q2_0.gguf + sha256: c2893fa6f951daee8b7263d5f49b455e57989ef56417a22125b03566dc560650 + uri: https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf/resolve/683f7be2629342a50987627b40881e54761a2430/Ternary-Bonsai-2-2B-Q2_0.gguf + - filename: bonsai/mmproj/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + sha256: ac9934e8d7966d0d62442c202ec8fdad99a045dea79366c1a4a246a46f5c0533 + uri: https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf/resolve/683f7be2629342a50987627b40881e54761a2430/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + overrides: + backend: bonsai + context_size: 32768 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: bonsai/mmproj/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + options: + - use_jinja:true + parameters: + min_p: 0.0 + model: bonsai/models/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-Q2_0.gguf + presence_penalty: 1.0 + repeat_penalty: 1.0 + temperature: 1.0 + top_k: 20 + top_p: 0.95 + template: + use_tokenizer_template: true - url: github:mudler/LocalAI/gallery/virtual.yaml@master name: adavar-3b-q4 description: | diff --git a/gallery/models/index.yaml b/gallery/models/index.yaml index da124ab3a..60f7cfbfe 100644 --- a/gallery/models/index.yaml +++ b/gallery/models/index.yaml @@ -4,6 +4,174 @@ # 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: ternary-bonsai-2-2b + name: ternary-bonsai-2-2b + version: 0.0.0-gallery.de98a2203208 + description: | + Ternary Bonsai 2 2B is a vision-language model derived from Qwen3.5-2B + with ternary transformer weights. This PQ2_0 build uses the Bonsai + backend and includes the Q8_0 vision projector. The three standard + packings contain the same weights in different storage layouts. + license: apache-2.0 + urls: + - https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf + - https://huggingface.co/Qwen/Qwen3.5-2B + tags: + - llm + - gguf + - reasoning + - vision + - multimodal + channel: gallery + policy: + requireSignature: false + schemaVersion: 1 + model: + url: github:mudler/LocalAI/gallery/virtual.yaml@master + files: + - filename: bonsai/models/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-PQ2_0.gguf + sha256: d5f01a4ecaf49239653308ee732ad1a0c5f802c0ddb997549fc7b694c8b45928 + uri: https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf/resolve/683f7be2629342a50987627b40881e54761a2430/Ternary-Bonsai-2-2B-PQ2_0.gguf + - filename: bonsai/mmproj/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + sha256: ac9934e8d7966d0d62442c202ec8fdad99a045dea79366c1a4a246a46f5c0533 + uri: https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf/resolve/683f7be2629342a50987627b40881e54761a2430/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + overrides: + backend: bonsai + context_size: 32768 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: bonsai/mmproj/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + options: + - use_jinja:true + parameters: + min_p: 0.0 + model: bonsai/models/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-PQ2_0.gguf + presence_penalty: 1.0 + repeat_penalty: 1.0 + temperature: 1.0 + top_k: 20 + top_p: 0.95 + template: + use_tokenizer_template: true + variants: + - model: ternary-bonsai-2-2b-ptq1 + - model: ternary-bonsai-2-2b-q2 +- kind: model + id: ternary-bonsai-2-2b-ptq1 + name: ternary-bonsai-2-2b-ptq1 + version: 0.0.0-gallery.a0b5119034e1 + description: | + Ternary Bonsai 2 2B is a vision-language model derived from Qwen3.5-2B + with ternary transformer weights. This PTQ1_0 build uses the Bonsai + backend and includes the Q8_0 vision projector. The three standard + packings contain the same weights in different storage layouts. + license: apache-2.0 + urls: + - https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf + - https://huggingface.co/Qwen/Qwen3.5-2B + tags: + - llm + - gguf + - reasoning + - vision + - multimodal + channel: gallery + policy: + requireSignature: false + schemaVersion: 1 + model: + url: github:mudler/LocalAI/gallery/virtual.yaml@master + files: + - filename: bonsai/models/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-PTQ1_0.gguf + sha256: 36061ad406df8a4903a6db005be350c50fabfaf717f9ef0acf4b76adcd9929fa + uri: https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf/resolve/683f7be2629342a50987627b40881e54761a2430/Ternary-Bonsai-2-2B-PTQ1_0.gguf + - filename: bonsai/mmproj/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + sha256: ac9934e8d7966d0d62442c202ec8fdad99a045dea79366c1a4a246a46f5c0533 + uri: https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf/resolve/683f7be2629342a50987627b40881e54761a2430/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + overrides: + backend: bonsai + context_size: 32768 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: bonsai/mmproj/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + options: + - use_jinja:true + parameters: + min_p: 0.0 + model: bonsai/models/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-PTQ1_0.gguf + presence_penalty: 1.0 + repeat_penalty: 1.0 + temperature: 1.0 + top_k: 20 + top_p: 0.95 + template: + use_tokenizer_template: true +- kind: model + id: ternary-bonsai-2-2b-q2 + name: ternary-bonsai-2-2b-q2 + version: 0.0.0-gallery.762a52ed03ee + description: | + Ternary Bonsai 2 2B is a vision-language model derived from Qwen3.5-2B + with ternary transformer weights. This Q2_0 build uses the Bonsai + backend and includes the Q8_0 vision projector. The three standard + packings contain the same weights in different storage layouts. + license: apache-2.0 + urls: + - https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf + - https://huggingface.co/Qwen/Qwen3.5-2B + tags: + - llm + - gguf + - reasoning + - vision + - multimodal + channel: gallery + policy: + requireSignature: false + schemaVersion: 1 + model: + url: github:mudler/LocalAI/gallery/virtual.yaml@master + files: + - filename: bonsai/models/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-Q2_0.gguf + sha256: c2893fa6f951daee8b7263d5f49b455e57989ef56417a22125b03566dc560650 + uri: https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf/resolve/683f7be2629342a50987627b40881e54761a2430/Ternary-Bonsai-2-2B-Q2_0.gguf + - filename: bonsai/mmproj/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + sha256: ac9934e8d7966d0d62442c202ec8fdad99a045dea79366c1a4a246a46f5c0533 + uri: https://huggingface.co/prism-ml/Ternary-Bonsai-2-2B-gguf/resolve/683f7be2629342a50987627b40881e54761a2430/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + overrides: + backend: bonsai + context_size: 32768 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: bonsai/mmproj/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-mmproj-Q8_0.gguf + options: + - use_jinja:true + parameters: + min_p: 0.0 + model: bonsai/models/ternary-bonsai-2-2b/Ternary-Bonsai-2-2B-Q2_0.gguf + presence_penalty: 1.0 + repeat_penalty: 1.0 + temperature: 1.0 + top_k: 20 + top_p: 0.95 + template: + use_tokenizer_template: true - kind: model id: adavar-3b-q4 name: adavar-3b-q4