@bendyline/gilde 0.1.18 → 0.1.20

package/authoring/chat-models/qwen3.6-27b-q4.json

@@ -0,0 +1,66 @@
+{
+  "id": "qwen3.6-27b-q4",
+  "name": "Qwen 3.6 (27B, Q4)",
+  "description": "Alibaba's Qwen 3.6 at 27B parameters. Multimodal (text + image + video), 256K native context, explicit native tool-calling support. Capable-tier on-device option for machines with 24GB+ memory.",
+  "tags": [
+    "alibaba",
+    "multimodal",
+    "vision",
+    "video",
+    "tools",
+    "long-context"
+  ],
+  "category": "general",
+  "maintainer": {
+    "name": "Alibaba",
+    "url": "https://ollama.com/library/qwen3.6"
+  },
+  "version": "1.1.4",
+  "updatedAt": "2026-07-28T00:00:00Z",
+  "license": "Apache-2.0",
+  "recoScore": 20,
+  "parameterSize": "27B",
+  "approxSizeBytes": 17106773120,
+  "supportsTools": true,
+  "contextWindow": 262144,
+  "upstream": "https://ollama.com/library/qwen3.6",
+  "style": {
+    "family": "qwen",
+    "reasoningFormat": "think",
+    "toolCallFormat": "function-call"
+  },
+  "behaviors": [
+    "reasoning.strip-think-tags",
+    "mcp.compact-tool-schemas",
+    "prompt.tool-cookbook-condensed",
+    "fabrication.detect-past-tense-no-tools",
+    "turn.ollama-num-predict-bumped",
+    "turn.preamble-folding",
+    "turn.ramble-detection"
+  ],
+  "ollama": {
+    "tag": "qwen3.6"
+  },
+  "llamaCpp": {
+    "huggingfaceRepo": "unsloth/Qwen3.6-27B-MTP-GGUF",
+    "filename": "Qwen3.6-27B-Q4_K_M.gguf",
+    "quantization": "Q4_K_M",
+    "residentBytes": 20528127744
+  },
+  "releasePatch": {
+    "tuning": {
+      "engine": {
+        "llamaCpp": {
+          "spec": {
+            "mtp": true,
+            "nMax": 4
+          }
+        }
+      }
+    }
+  },
+  "mlx": {
+    "huggingfaceRepo": "mlx-community/Qwen3.6-27B-4bit",
+    "quantization": "4bit"
+  }
+}

package/authoring/chat-models/qwen3.6-27b-q8.json

@@ -0,0 +1,153 @@
+{
+  "id": "qwen3.6-27b-q8",
+  "name": "Qwen 3.6 (27B, Q8)",
+  "description": "Alibaba's Qwen 3.6 at 27B parameters, 8-bit quantization for higher fidelity than the Q4 build (at roughly double the memory footprint). Multimodal (text + image + video), 256K native context, explicit native tool-calling support. For machines with ~40GB+ memory.",
+  "tags": [
+    "alibaba",
+    "multimodal",
+    "vision",
+    "video",
+    "tools",
+    "long-context"
+  ],
+  "category": "general",
+  "maintainer": {
+    "name": "Alibaba",
+    "url": "https://ollama.com/library/qwen3.6"
+  },
+  "version": "1.0.2",
+  "updatedAt": "2026-07-28T00:00:00Z",
+  "license": "Apache-2.0",
+  "recoScore": 20,
+  "parameterSize": "27B",
+  "approxSizeBytes": 29047084160,
+  "supportsTools": true,
+  "contextWindow": 262144,
+  "upstream": "https://ollama.com/library/qwen3.6",
+  "style": {
+    "family": "qwen",
+    "reasoningFormat": "think",
+    "toolCallFormat": "function-call"
+  },
+  "behaviors": [
+    "reasoning.strip-think-tags",
+    "mcp.compact-tool-schemas",
+    "prompt.tool-cookbook-condensed",
+    "fabrication.detect-past-tense-no-tools",
+    "turn.ollama-num-predict-bumped",
+    "turn.preamble-folding",
+    "turn.ramble-detection"
+  ],
+  "tuning": {
+    "sampling": {
+      "temperature": 0.7,
+      "topP": 0.8,
+      "topK": 20,
+      "minP": 0,
+      "maxTokens": 8192
+    },
+    "samplingWhenThinking": {
+      "temperature": 0.6,
+      "topP": 0.95
+    },
+    "reasoning": {
+      "enableThinking": true,
+      "thinkingBudget": 4096
+    },
+    "promptTags": {
+      "enableThinkingTag": "/think",
+      "disableThinkingTag": "/no_think"
+    },
+    "profiles": {
+      "thinking-general": {
+        "sampling": {
+          "temperature": 0.8,
+          "topP": 0.95,
+          "topK": 20,
+          "minP": 0,
+          "repetitionPenalty": 1,
+          "maxTokens": 12288
+        },
+        "reasoning": {
+          "enableThinking": true,
+          "thinkingBudget": 4096
+        }
+      },
+      "thinking-coding": {
+        "sampling": {
+          "temperature": 0.6,
+          "topP": 0.95,
+          "topK": 20,
+          "minP": 0,
+          "repetitionPenalty": 1,
+          "maxTokens": 16384
+        },
+        "reasoning": {
+          "enableThinking": true,
+          "thinkingBudget": 6144
+        }
+      },
+      "thinking-precise": {
+        "sampling": {
+          "temperature": 0.4,
+          "topP": 0.9,
+          "topK": 20,
+          "minP": 0,
+          "seed": 0,
+          "maxTokens": 6144
+        },
+        "reasoning": {
+          "enableThinking": true,
+          "thinkingBudget": 4096
+        }
+      },
+      "instruct": {
+        "sampling": {
+          "temperature": 0.7,
+          "topP": 0.8,
+          "topK": 20,
+          "minP": 0,
+          "presencePenalty": 1.5,
+          "repetitionPenalty": 1,
+          "maxTokens": 4096
+        },
+        "reasoning": {
+          "enableThinking": false
+        }
+      },
+      "creative": {
+        "sampling": {
+          "temperature": 1.1,
+          "topP": 0.95,
+          "topK": 40,
+          "maxTokens": 8192
+        },
+        "reasoning": {
+          "enableThinking": false
+        }
+      }
+    }
+  },
+  "llamaCpp": {
+    "huggingfaceRepo": "unsloth/Qwen3.6-27B-MTP-GGUF",
+    "filename": "Qwen3.6-27B-Q8_0.gguf",
+    "quantization": "Q8_0",
+    "residentBytes": 34856500992
+  },
+  "releasePatch": {
+    "tuning": {
+      "engine": {
+        "llamaCpp": {
+          "spec": {
+            "mtp": true,
+            "nMax": 4
+          }
+        }
+      }
+    }
+  },
+  "mlx": {
+    "huggingfaceRepo": "mlx-community/Qwen3.6-27B-8bit",
+    "quantization": "8bit"
+  }
+}

package/authoring/chat-models/qwen3.6-35b-a3b-q4.json

@@ -0,0 +1,154 @@
+{
+  "id": "qwen3.6-35b-a3b-q4",
+  "name": "Qwen 3.6 (35B-A3B, Q4)",
+  "description": "Alibaba's Qwen 3.6 at 35B total parameters with only 3B active per token (Mixture-of-Experts: 256 experts, 8 routed + 1 shared). Multimodal (text + image + video), 256K native context (extensible to ~1M with YaRN), explicit native tool-calling support. MoE sparsity keeps inference light despite the larger total parameter count.",
+  "tags": [
+    "alibaba",
+    "multimodal",
+    "vision",
+    "video",
+    "tools",
+    "long-context",
+    "moe"
+  ],
+  "category": "general",
+  "maintainer": {
+    "name": "Alibaba",
+    "url": "https://huggingface.co/Qwen/Qwen3.6-35B-A3B"
+  },
+  "version": "1.0.1",
+  "updatedAt": "2026-07-28T00:00:00Z",
+  "license": "Apache-2.0",
+  "recoScore": 20,
+  "parameterSize": "35B",
+  "approxSizeBytes": 22663387424,
+  "supportsTools": true,
+  "contextWindow": 262144,
+  "upstream": "https://huggingface.co/Qwen/Qwen3.6-35B-A3B",
+  "style": {
+    "family": "qwen",
+    "reasoningFormat": "think",
+    "toolCallFormat": "function-call"
+  },
+  "behaviors": [
+    "reasoning.strip-think-tags",
+    "mcp.compact-tool-schemas",
+    "prompt.tool-cookbook-condensed",
+    "fabrication.detect-past-tense-no-tools",
+    "turn.ollama-num-predict-bumped",
+    "turn.preamble-folding",
+    "turn.ramble-detection"
+  ],
+  "tuning": {
+    "sampling": {
+      "temperature": 0.7,
+      "topP": 0.8,
+      "topK": 20,
+      "minP": 0,
+      "maxTokens": 8192
+    },
+    "samplingWhenThinking": {
+      "temperature": 0.6,
+      "topP": 0.95
+    },
+    "reasoning": {
+      "enableThinking": true,
+      "thinkingBudget": 4096
+    },
+    "promptTags": {
+      "enableThinkingTag": "/think",
+      "disableThinkingTag": "/no_think"
+    },
+    "profiles": {
+      "thinking-general": {
+        "sampling": {
+          "temperature": 0.8,
+          "topP": 0.95,
+          "topK": 20,
+          "minP": 0,
+          "repetitionPenalty": 1,
+          "maxTokens": 12288
+        },
+        "reasoning": {
+          "enableThinking": true,
+          "thinkingBudget": 4096
+        }
+      },
+      "thinking-coding": {
+        "sampling": {
+          "temperature": 0.6,
+          "topP": 0.95,
+          "topK": 20,
+          "minP": 0,
+          "repetitionPenalty": 1,
+          "maxTokens": 16384
+        },
+        "reasoning": {
+          "enableThinking": true,
+          "thinkingBudget": 6144
+        }
+      },
+      "thinking-precise": {
+        "sampling": {
+          "temperature": 0.4,
+          "topP": 0.9,
+          "topK": 20,
+          "minP": 0,
+          "seed": 0,
+          "maxTokens": 6144
+        },
+        "reasoning": {
+          "enableThinking": true,
+          "thinkingBudget": 4096
+        }
+      },
+      "instruct": {
+        "sampling": {
+          "temperature": 0.7,
+          "topP": 0.8,
+          "topK": 20,
+          "minP": 0,
+          "presencePenalty": 1.5,
+          "repetitionPenalty": 1,
+          "maxTokens": 4096
+        },
+        "reasoning": {
+          "enableThinking": false
+        }
+      },
+      "creative": {
+        "sampling": {
+          "temperature": 1.1,
+          "topP": 0.95,
+          "topK": 40,
+          "maxTokens": 8192
+        },
+        "reasoning": {
+          "enableThinking": false
+        }
+      }
+    }
+  },
+  "llamaCpp": {
+    "huggingfaceRepo": "unsloth/Qwen3.6-35B-A3B-MTP-GGUF",
+    "filename": "Qwen3.6-35B-A3B-UD-Q4_K_M.gguf",
+    "quantization": "UD-Q4_K_M",
+    "residentBytes": 27196064909
+  },
+  "releasePatch": {
+    "tuning": {
+      "engine": {
+        "llamaCpp": {
+          "spec": {
+            "mtp": true,
+            "nMax": 4
+          }
+        }
+      }
+    }
+  },
+  "mlx": {
+    "huggingfaceRepo": "mlx-community/Qwen3.6-35B-A3B-4bit",
+    "quantization": "4bit"
+  }
+}

package/authoring/chat-models/qwen3.6-35b-a3b-q8.json

@@ -0,0 +1,154 @@
+{
+  "id": "qwen3.6-35b-a3b-q8",
+  "name": "Qwen 3.6 (35B-A3B, Q8)",
+  "description": "Alibaba's Qwen 3.6 at 35B total parameters with only 3B active per token (Mixture-of-Experts: 256 experts, 8 routed + 1 shared), 8-bit quantization for higher fidelity than the Q4 build (at roughly double the memory footprint). Multimodal (text + image + video), 256K native context (extensible to ~1M with YaRN), explicit native tool-calling support.",
+  "tags": [
+    "alibaba",
+    "multimodal",
+    "vision",
+    "video",
+    "tools",
+    "long-context",
+    "moe"
+  ],
+  "category": "general",
+  "maintainer": {
+    "name": "Alibaba",
+    "url": "https://huggingface.co/Qwen/Qwen3.6-35B-A3B"
+  },
+  "version": "1.0.1",
+  "updatedAt": "2026-07-28T00:00:00Z",
+  "license": "Apache-2.0",
+  "recoScore": 20,
+  "parameterSize": "35B",
+  "approxSizeBytes": 37801097504,
+  "supportsTools": true,
+  "contextWindow": 262144,
+  "upstream": "https://huggingface.co/Qwen/Qwen3.6-35B-A3B",
+  "style": {
+    "family": "qwen",
+    "reasoningFormat": "think",
+    "toolCallFormat": "function-call"
+  },
+  "behaviors": [
+    "reasoning.strip-think-tags",
+    "mcp.compact-tool-schemas",
+    "prompt.tool-cookbook-condensed",
+    "fabrication.detect-past-tense-no-tools",
+    "turn.ollama-num-predict-bumped",
+    "turn.preamble-folding",
+    "turn.ramble-detection"
+  ],
+  "tuning": {
+    "sampling": {
+      "temperature": 0.7,
+      "topP": 0.8,
+      "topK": 20,
+      "minP": 0,
+      "maxTokens": 8192
+    },
+    "samplingWhenThinking": {
+      "temperature": 0.6,
+      "topP": 0.95
+    },
+    "reasoning": {
+      "enableThinking": true,
+      "thinkingBudget": 4096
+    },
+    "promptTags": {
+      "enableThinkingTag": "/think",
+      "disableThinkingTag": "/no_think"
+    },
+    "profiles": {
+      "thinking-general": {
+        "sampling": {
+          "temperature": 0.8,
+          "topP": 0.95,
+          "topK": 20,
+          "minP": 0,
+          "repetitionPenalty": 1,
+          "maxTokens": 12288
+        },
+        "reasoning": {
+          "enableThinking": true,
+          "thinkingBudget": 4096
+        }
+      },
+      "thinking-coding": {
+        "sampling": {
+          "temperature": 0.6,
+          "topP": 0.95,
+          "topK": 20,
+          "minP": 0,
+          "repetitionPenalty": 1,
+          "maxTokens": 16384
+        },
+        "reasoning": {
+          "enableThinking": true,
+          "thinkingBudget": 6144
+        }
+      },
+      "thinking-precise": {
+        "sampling": {
+          "temperature": 0.4,
+          "topP": 0.9,
+          "topK": 20,
+          "minP": 0,
+          "seed": 0,
+          "maxTokens": 6144
+        },
+        "reasoning": {
+          "enableThinking": true,
+          "thinkingBudget": 4096
+        }
+      },
+      "instruct": {
+        "sampling": {
+          "temperature": 0.7,
+          "topP": 0.8,
+          "topK": 20,
+          "minP": 0,
+          "presencePenalty": 1.5,
+          "repetitionPenalty": 1,
+          "maxTokens": 4096
+        },
+        "reasoning": {
+          "enableThinking": false
+        }
+      },
+      "creative": {
+        "sampling": {
+          "temperature": 1.1,
+          "topP": 0.95,
+          "topK": 40,
+          "maxTokens": 8192
+        },
+        "reasoning": {
+          "enableThinking": false
+        }
+      }
+    }
+  },
+  "llamaCpp": {
+    "huggingfaceRepo": "unsloth/Qwen3.6-35B-A3B-MTP-GGUF",
+    "filename": "Qwen3.6-35B-A3B-Q8_0.gguf",
+    "quantization": "Q8_0",
+    "residentBytes": 45361317005
+  },
+  "releasePatch": {
+    "tuning": {
+      "engine": {
+        "llamaCpp": {
+          "spec": {
+            "mtp": true,
+            "nMax": 4
+          }
+        }
+      }
+    }
+  },
+  "mlx": {
+    "huggingfaceRepo": "mlx-community/Qwen3.6-35B-A3B-8bit",
+    "quantization": "8bit"
+  }
+}

package/authoring/chat-models/talkie-1930-13b-q4.json

@@ -0,0 +1,74 @@
+{
+  "id": "talkie-1930-13b-q4",
+  "name": "Talkie 1930 (13B)",
+  "description": "A 13B instruct model trained solely on pre-1931 English-language text — it writes and converses in an authentic period voice, free of modern anachronisms. Best for creative writing, correspondence, and in-character chat rather than tool use or current-events questions. Runs a community GGUF that rescales talkie's out-of-range layer-14 ffn_down channel so K-quantization stays coherent. 2K context. Apache-2.0.",
+  "tags": [
+    "talkie-lm",
+    "creative",
+    "writing",
+    "vintage",
+    "english"
+  ],
+  "category": "general",
+  "maintainer": {
+    "name": "talkie-lm",
+    "url": "https://huggingface.co/talkie-lm/talkie-1930-13b-it"
+  },
+  "version": "1.0.0",
+  "updatedAt": "2026-07-20T00:00:00Z",
+  "license": "Apache-2.0",
+  "licenseClass": "open",
+  "licenseShortName": "Apache 2.0",
+  "parameterSize": "13B",
+  "approxSizeBytes": 8572111136,
+  "supportsTools": false,
+  "contextWindow": 2048,
+  "upstream": "https://huggingface.co/talkie-lm/talkie-1930-13b-it",
+  "style": {
+    "family": "other",
+    "reasoningFormat": "none",
+    "toolCallFormat": "function-call"
+  },
+  "behaviors": [
+    "prompt.minimal-context",
+    "turn.preamble-folding",
+    {
+      "id": "turn.ramble-detection",
+      "config": {
+        "coldThreshold": 4000,
+        "postActionThreshold": 1200
+      }
+    }
+  ],
+  "tuning": {
+    "sampling": {
+      "temperature": 0.8,
+      "topP": 0.95,
+      "topK": 40,
+      "repetitionPenalty": 1.1,
+      "repetitionContext": 64,
+      "maxTokens": 1024
+    },
+    "reasoning": {
+      "enableThinking": false
+    },
+    "profiles": {
+      "creative": {
+        "sampling": {
+          "temperature": 1,
+          "topP": 0.95,
+          "topK": 60,
+          "repetitionPenalty": 1.05,
+          "repetitionContext": 32,
+          "maxTokens": 1024
+        }
+      }
+    }
+  },
+  "llamaCpp": {
+    "huggingfaceRepo": "zakarth/talkie-1930-13b-it-vulkan-fixed-GGUF",
+    "filename": "talkie-1930-13b-it.Q4_K_M.gguf",
+    "quantization": "Q4_K_M",
+    "residentBytes": 10286533363
+  }
+}