diff --git a/gallery/index.yaml b/gallery/index.yaml index 4d8fe4d31a46..d798524714ea 100644 --- a/gallery/index.yaml +++ b/gallery/index.yaml @@ -5549,6 +5549,8 @@ variants: - model: laguna-s-2.1-q8 - model: laguna-s-2.1-dflash + - model: laguna-s-2.1-apex-i-quality + - model: laguna-s-2.1-apex-i-compact description: | Laguna S 2.1 is Poolside's 118B-parameter, 8B-active Mixture-of-Experts model for agentic software engineering. It supports tool use and a native one-million-token context window; the official GGUF recommends 256K context for best output quality. This default entry uses the current 96 GB Q4_K_M artifact, with imatrix-quantized routed experts and a Q8_0 signal path. @@ -5572,6 +5574,70 @@ - filename: llama-cpp/models/Laguna-S-2.1-GGUF/laguna-s-2.1-Q4_K_M.gguf sha256: a8b55c75714ea73fd90ec85de5defdc0b8d88ca0ad2108343cdd8fc22f7583e4 uri: https://huggingface.co/poolside/Laguna-S-2.1-GGUF/resolve/main/laguna-s-2.1-Q4_K_M.gguf +- !!merge <<: *laguna-s-2-1-q8 + name: "laguna-s-2.1-apex-i-quality" + variants: [] + urls: + - https://huggingface.co/poolside/Laguna-S-2.1 + - https://huggingface.co/Myric/Laguna-S-2.1-APEX-GGUF + description: | + Laguna S 2.1 in the 73.9 GB APEX-I Quality format. This community build + uses an importance matrix and mixed precision to preserve the always-active + signal path while reducing the memory required by the routed experts. + + License: OpenMDW 1.1. + last_checked: "2026-08-26" + overrides: + backend: llama-cpp + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - completion + options: + - use_jinja:true + parameters: + model: llama-cpp/models/Laguna-S-2.1-APEX-GGUF/Laguna-S-2.1-APEX-i-quality-v2.gguf + template: + use_tokenizer_template: true + files: + - filename: llama-cpp/models/Laguna-S-2.1-APEX-GGUF/Laguna-S-2.1-APEX-i-quality-v2.gguf + sha256: 772886614b2f11ba62ea2c2ebb5d02804e1ab829329b50f644e83e174decfaf2 + uri: huggingface://Myric/Laguna-S-2.1-APEX-GGUF/Laguna-S-2.1-APEX-i-quality-v2.gguf +- !!merge <<: *laguna-s-2-1-q8 + name: "laguna-s-2.1-apex-i-compact" + variants: [] + urls: + - https://huggingface.co/poolside/Laguna-S-2.1 + - https://huggingface.co/Myric/Laguna-S-2.1-APEX-GGUF + description: | + Laguna S 2.1 in the smaller 54.4 GB APEX-I Compact format. This community + build uses an importance matrix and mixed precision to reduce memory use + while retaining higher precision for the always-active signal path. + + License: OpenMDW 1.1. + last_checked: "2026-08-26" + overrides: + backend: llama-cpp + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - completion + options: + - use_jinja:true + parameters: + model: llama-cpp/models/Laguna-S-2.1-APEX-GGUF/Laguna-S-2.1-APEX-i-compact-v2.gguf + template: + use_tokenizer_template: true + files: + - filename: llama-cpp/models/Laguna-S-2.1-APEX-GGUF/Laguna-S-2.1-APEX-i-compact-v2.gguf + sha256: b5964de8c5de37c9d4fcbd6894a68a95a81a2c404b12897d9581fc0d934134de + uri: huggingface://Myric/Laguna-S-2.1-APEX-GGUF/Laguna-S-2.1-APEX-i-compact-v2.gguf - !!merge <<: *laguna-s-2-1-q8 name: "laguna-s-2.1-dflash" description: |