From deb430f3ecec3a726ed8a377ef9a1aeef2f2747e Mon Sep 17 00:00:00 2001 From: "LocalAI [bot]" <139863280+localai-bot@users.noreply.github.com> Date: Tue, 23 Jun 2026 23:15:47 +0200 Subject: [PATCH] chore(model-gallery): :arrow_up: update checksum (#10469) :arrow_up: Checksum updates in gallery/index.yaml Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> --- gallery/index.yaml | 97 +++------------------------------------------- 1 file changed, 5 insertions(+), 92 deletions(-) diff --git a/gallery/index.yaml b/gallery/index.yaml index 29f284ad8..255e828f5 100644 --- a/gallery/index.yaml +++ b/gallery/index.yaml @@ -3,28 +3,7 @@ url: "github:mudler/LocalAI/gallery/virtual.yaml@master" urls: - https://huggingface.co/Jackrong/Qwopus3.6-27B-Coder-Compat-MTP-GGUF - description: | - ๐Ÿช Qwopus-3.6-27B-Coder - Coder SFT Release - - Agentic Coding & Tool-Use Reasoning Model Fine-Tuned on Qwopus3.6-27B-v2 - - ๐Ÿงฌ Trace Inversion & Negentropy - ๐Ÿง  27B Dense Model - โšก Agentic Coding - ๐Ÿ› ๏ธ Tool Calling & Agent - ๐Ÿ† SWE-bench Verified: 67.0% (off-thinking) - - ๐Ÿ’ก What is Qwopus-3.6-27B-Coder? - ๐Ÿช Qwopus-3.6-27B-Coder is a reasoning-enhanced agentic coding model built on top of Qwopus3.6-27B-v2. It inherits the powerful reasoning foundation of the v2 base โ€” which achieved 87.43% MMLU-Pro and 75.25% SWE-bench Verified โ€” and further specializes it for agentic code generation, structured tool calling, debugging, and instruction-following in developer workflows. The model is designed to excel at repository-level coding tasks, multi-turn tool orchestration, and complex logical reasoning under realistic agent environments. - - ๐Ÿงฉ Agentic Coding - Optimized for repository-level coding, debugging, patch generation, and structured multi-step development workflows. - - ๐Ÿ› ๏ธ Tool Calling - Learns from real agent trajectories with tool definitions, tool calls, and environment feedback for robust multi-turn execution. - - ... + description: "\U0001FA90 Qwopus-3.6-27B-Coder\nCoder SFT Release\n\nAgentic Coding & Tool-Use Reasoning Model Fine-Tuned on Qwopus3.6-27B-v2\n\n\U0001F9EC Trace Inversion & Negentropy\n\U0001F9E0 27B Dense Model\nโšก Agentic Coding\n\U0001F6E0๏ธ Tool Calling & Agent\n\U0001F3C6 SWE-bench Verified: 67.0% (off-thinking)\n\n\U0001F4A1 What is Qwopus-3.6-27B-Coder?\n\U0001FA90 Qwopus-3.6-27B-Coder is a reasoning-enhanced agentic coding model built on top of Qwopus3.6-27B-v2. It inherits the powerful reasoning foundation of the v2 base โ€” which achieved 87.43% MMLU-Pro and 75.25% SWE-bench Verified โ€” and further specializes it for agentic code generation, structured tool calling, debugging, and instruction-following in developer workflows. The model is designed to excel at repository-level coding tasks, multi-turn tool orchestration, and complex logical reasoning under realistic agent environments.\n\n\U0001F9E9 Agentic Coding\nOptimized for repository-level coding, debugging, patch generation, and structured multi-step development workflows.\n\n\U0001F6E0๏ธ Tool Calling\nLearns from real agent trajectories with tool definitions, tool calls, and environment feedback for robust multi-turn execution.\n\n...\n" license: "apache-2.0" tags: - llm @@ -241,33 +220,7 @@ url: "github:mudler/LocalAI/gallery/virtual.yaml@master" urls: - https://huggingface.co/unsloth/GLM-5.2-GGUF - description: | - # GLM-5.2 - - ๐Ÿ‘‹ Join our WeChat or Discord community. - - ๐Ÿ“– Check out the GLM-5.2 blog and GLM-5 Technical report. - - ๐Ÿ“ Use GLM-5.2 API services on Z.ai API Platform. - - ๐Ÿ”œ Try GLM-5.2 here. - - [Paper] - [GitHub] - - ## Introduction - - We're introducing GLM-5.2, our latest flagship model for long-horizon tasks. It marks a substantial leap in long-horizon task capability over its predecessor GLM-5.1 and, for the first time, delivers that capability on a **solid 1M-token context**. GLM-5.2's new capabilities include: - - **Solid 1M Context:** A solid 1M-token context that stably sustains long-horizon work - - **Advanced Coding with Flexible Effort**: Stronger coding capabilities with multiple thinking effort levels to balance performance and latency - - **Improved Architecture**: We propose IndexShare, which reuses the same indexer across every four sparse attention layers, reducing per-token FLOPs by 2.9ร— at a 1M context length. We also improve GLM-5.2โ€™s MTP layer for speculative decoding, increasing the acceptance length by up to 20% - - **Pure Open**: An MIT open-source license โ€” no regional limits, technical access without borders - - ## Benchmark - - ## Serve GLM-5.2 Locally - - ... + description: "# GLM-5.2\n\n\U0001F44B Join our WeChat or Discord community.\n\n\U0001F4D6 Check out the GLM-5.2 blog and GLM-5 Technical report.\n\n\U0001F4CD Use GLM-5.2 API services on Z.ai API Platform.\n\n\U0001F51C Try GLM-5.2 here.\n\n[Paper]\n[GitHub]\n\n## Introduction\n\nWe're introducing GLM-5.2, our latest flagship model for long-horizon tasks. It marks a substantial leap in long-horizon task capability over its predecessor GLM-5.1 and, for the first time, delivers that capability on a **solid 1M-token context**. GLM-5.2's new capabilities include:\n - **Solid 1M Context:** A solid 1M-token context that stably sustains long-horizon work\n - **Advanced Coding with Flexible Effort**: Stronger coding capabilities with multiple thinking effort levels to balance performance and latency\n - **Improved Architecture**: We propose IndexShare, which reuses the same indexer across every four sparse attention layers, reducing per-token FLOPs by 2.9ร— at a 1M context length. We also improve GLM-5.2โ€™s MTP layer for speculative decoding, increasing the acceptance length by up to 20%\n - **Pure Open**: An MIT open-source license โ€” no regional limits, technical access without borders\n\n## Benchmark\n\n## Serve GLM-5.2 Locally\n\n...\n" license: "mit" tags: - llm @@ -390,26 +343,7 @@ url: "github:mudler/LocalAI/gallery/virtual.yaml@master" urls: - https://huggingface.co/michaelw9999/Qwopus3.6-27B-v2-MTP-NVFP4-GGUF - description: | - ๐Ÿช Qwopus3.6-27B-v2-MTP - MTP Release - - Multi-Token Prediction reasoning model fine-tuned from Qwen3.6-27B - - ๐Ÿงฌ Trace Inversion & Negentropy - ๐Ÿง  27B Parameters - โšก Speculative Decoding - ๐Ÿ› ๏ธ Coding / DevOps / Math - - ๐Ÿ’ก What is Qwopus3.6-27B-v2-MTP? - ๐Ÿช Qwopus3.6-27B-v2-MTP is a speed-oriented reasoning release built on top of Qwen3.6-27B. It keeps the Qwopus line's focus on reconstructed reasoning traces, coding discipline, DevOps procedures, and mathematical derivations, while adding Multi-Token Prediction for faster generation. The goal is simple: preserve the depth and structure of a 27B reasoning model while making real interactive use noticeably faster. - - โšก MTP DecodingAuxiliary future-token prediction improves throughput on long reasoning, code, math, and strict-format prompts. - ๐Ÿงฉ Structured ReasoningInherits the Qwopus training recipe built around reconstructed step-by-step reasoning trajectories. - ๐Ÿงช GB10 TestedValidated on a 30-question local benchmark across Logic, Coding, DevOps, Math, and Edge tasks. - ๐Ÿš€ Practical SpeedDesigned for workflows where strong answers matter, but waiting several extra minutes per task does not. - - ... + description: "\U0001FA90 Qwopus3.6-27B-v2-MTP\nMTP Release\n\nMulti-Token Prediction reasoning model fine-tuned from Qwen3.6-27B\n\n\U0001F9EC Trace Inversion & Negentropy\n\U0001F9E0 27B Parameters\nโšก Speculative Decoding\n\U0001F6E0๏ธ Coding / DevOps / Math\n\n\U0001F4A1 What is Qwopus3.6-27B-v2-MTP?\n\U0001FA90 Qwopus3.6-27B-v2-MTP is a speed-oriented reasoning release built on top of Qwen3.6-27B. It keeps the Qwopus line's focus on reconstructed reasoning traces, coding discipline, DevOps procedures, and mathematical derivations, while adding Multi-Token Prediction for faster generation. The goal is simple: preserve the depth and structure of a 27B reasoning model while making real interactive use noticeably faster.\n\nโšก MTP DecodingAuxiliary future-token prediction improves throughput on long reasoning, code, math, and strict-format prompts.\n\U0001F9E9 Structured ReasoningInherits the Qwopus training recipe built around reconstructed step-by-step reasoning trajectories.\n\U0001F9EA GB10 TestedValidated on a 30-question local benchmark across Logic, Coding, DevOps, Math, and Edge tasks.\n\U0001F680 Practical SpeedDesigned for workflows where strong answers matter, but waiting several extra minutes per task does not.\n\n...\n" tags: - llm - gguf @@ -435,28 +369,7 @@ url: "github:mudler/LocalAI/gallery/virtual.yaml@master" urls: - https://huggingface.co/michaelw9999/Qwopus3.6-27B-Coder-MTP-NVFP4-GGUF - description: | - ๐Ÿช Qwopus-3.6-27B-Coder - Coder SFT Release - - Agentic Coding & Tool-Use Reasoning Model Fine-Tuned on Qwopus3.6-27B-v2 - - ๐Ÿงฌ Trace Inversion & Negentropy - ๐Ÿง  27B Dense Model - โšก Agentic Coding - ๐Ÿ› ๏ธ Tool Calling & Agent - ๐Ÿ† SWE-bench Verified: 67.0% (off-thinking) - - ๐Ÿ’ก What is Qwopus-3.6-27B-Coder? - ๐Ÿช Qwopus-3.6-27B-Coder is a reasoning-enhanced agentic coding model built on top of Qwopus3.6-27B-v2. It inherits the powerful reasoning foundation of the v2 base โ€” which achieved 87.43% MMLU-Pro (300ex) and 75.25% SWE-bench Verified โ€” and further specializes it for agentic code generation, structured tool calling, debugging, and instruction-following in developer workflows. The model is designed to excel at repository-level coding tasks, multi-turn tool orchestration, and complex logical reasoning under realistic agent environments. - - ๐Ÿงฉ Agentic Coding - Optimized for repository-level coding, debugging, patch generation, and structured multi-step development workflows. - - ๐Ÿ› ๏ธ Tool Calling - Learns from real agent trajectories with tool definitions, tool calls, and environment feedback for robust multi-turn execution. - - ... + description: "\U0001FA90 Qwopus-3.6-27B-Coder\nCoder SFT Release\n\nAgentic Coding & Tool-Use Reasoning Model Fine-Tuned on Qwopus3.6-27B-v2\n\n\U0001F9EC Trace Inversion & Negentropy\n\U0001F9E0 27B Dense Model\nโšก Agentic Coding\n\U0001F6E0๏ธ Tool Calling & Agent\n\U0001F3C6 SWE-bench Verified: 67.0% (off-thinking)\n\n\U0001F4A1 What is Qwopus-3.6-27B-Coder?\n\U0001FA90 Qwopus-3.6-27B-Coder is a reasoning-enhanced agentic coding model built on top of Qwopus3.6-27B-v2. It inherits the powerful reasoning foundation of the v2 base โ€” which achieved 87.43% MMLU-Pro (300ex) and 75.25% SWE-bench Verified โ€” and further specializes it for agentic code generation, structured tool calling, debugging, and instruction-following in developer workflows. The model is designed to excel at repository-level coding tasks, multi-turn tool orchestration, and complex logical reasoning under realistic agent environments.\n\n\U0001F9E9 Agentic Coding\nOptimized for repository-level coding, debugging, patch generation, and structured multi-step development workflows.\n\n\U0001F6E0๏ธ Tool Calling\nLearns from real agent trajectories with tool definitions, tool calls, and environment feedback for robust multi-turn execution.\n\n...\n" tags: - llm - gguf @@ -1676,8 +1589,8 @@ use_tokenizer_template: true files: - filename: llama-cpp/models/Qwopus3.6-27B-v2-MTP-GGUF/Qwopus3.6-27B-v2-MTP-Q4_K_M.gguf - sha256: 818d68223be4d8518dac0b3b5604dde633cbbcbae1f491d842a3e26711c6606d uri: https://huggingface.co/Jackrong/Qwopus3.6-27B-v2-MTP-GGUF/resolve/main/Qwopus3.6-27B-v2-MTP-Q4_K_M.gguf + sha256: 31cf5fc2406a0c7aaebcc26d440bf0df94e215d0589d5205bf319649c052b50a - name: "qwen3.6-40b-claude-4.6-opus-deckard-heretic-uncensored-thinking-neo-code-di-imatrix-max" url: "github:mudler/LocalAI/gallery/virtual.yaml@master" urls: