@bendyline/gilde 0.1.29 → 0.1.31

package/data/chat-models/ne/nemotron3.5-lightning-30b-q4/versions/1.0.1/manifest.json

@@ -0,0 +1,73 @@
+{
+  "schemaVersion": 1,
+  "version": "1.0.1",
+  "releasedAt": "2026-08-15T00:00:00Z",
+  "approxSizeBytes": 25430738944,
+  "llamaCpp": {
+    "huggingfaceRepo": "ggml-org/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-GGUF",
+    "revision": "88d7ce0b0fa385c5108866ce5d33690927531a37",
+    "filename": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Q4_K_M.gguf",
+    "sha256": "6110e2e2e6cd324e6ee69ddced5a6b34fad6c94ca9827222a1e420fb92e3c90b",
+    "approxSizeBytes": 25430738944,
+    "quantization": "Q4_K_M"
+  },
+  "mlx": {
+    "huggingfaceRepo": "mlx-community/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-4bit",
+    "revision": "55ac8c89261109b36c04371cd3f479a4594208c8",
+    "quantization": "4bit",
+    "approxSizeBytes": 17792581692,
+    "files": [
+      {
+        "name": "chat_template.jinja",
+        "sha256": "58933db77d3099b4f78c55a38347a72e1ea05b97d6bd8f38775303dc0194e0a9",
+        "sizeBytes": 9867
+      },
+      {
+        "name": "config.json",
+        "sha256": "2d4bdc543efd8ccd68fccbf1231b1d1216ab2a0482229ea5638d448127973b27",
+        "sizeBytes": 2932
+      },
+      {
+        "name": "generation_config.json",
+        "sha256": "0ed94823ae737b507aef6af01e837041b7aa2fb0306793a310cd201d89a31ff4",
+        "sizeBytes": 210
+      },
+      {
+        "name": "model-00001-of-00004.safetensors",
+        "sha256": "12e2ba8a7f0990b7e34ddb25cf4043fe39b105215fe61a0b7dc1447be6b247b3",
+        "sizeBytes": 5118631442
+      },
+      {
+        "name": "model-00002-of-00004.safetensors",
+        "sha256": "a095b11193a9d58e2c67c83bc2aff0eedde8f76b77ff8ee4781fe08f6c3f5363",
+        "sizeBytes": 5291568433
+      },
+      {
+        "name": "model-00003-of-00004.safetensors",
+        "sha256": "c55808e0cf33a6b957d34ab57caf4185c4aed49c6ef15c7c69683e7af7e892cc",
+        "sizeBytes": 5291568429
+      },
+      {
+        "name": "model-00004-of-00004.safetensors",
+        "sha256": "9c541daef892060fffb831afb9e20251ff18c492df7bacd0bfee4d8e43124d89",
+        "sizeBytes": 2073657055
+      },
+      {
+        "name": "model.safetensors.index.json",
+        "sha256": "4c12ae32746441ec65e94d8cf312462ac10ab0d3c80513bba3d1c1ca3f2ed9ca",
+        "sizeBytes": 65410
+      },
+      {
+        "name": "tokenizer.json",
+        "sha256": "623c34567aebb18582765289fbe23d901c62704d6518d71866e0e58db892b5b7",
+        "sizeBytes": 17077484
+      },
+      {
+        "name": "tokenizer_config.json",
+        "sha256": "ec380f0a6c0c2e294eb2435f5ec521fb2121cfc970e6ebf023d06829714a4eaa",
+        "sizeBytes": 430
+      }
+    ]
+  },
+  "minGezelVersion": "1.26227"
+}

package/data/chat-models/ne/nemotron3.5-lightning-30b-q8/manifest.json

@@ -3,7 +3,7 @@
   "kind": "chat-model",
   "id": "nemotron3.5-lightning-30b-q8",
   "name": "Nemotron 3.5 Lightning (30B-A3B, Q8)",
-  "description": "NVIDIA's Nemotron 3.5 Lightning — a text-only 30B-parameter hybrid Mamba-2/Transformer Mixture-of-Experts model with about 3B active parameters per token, native reasoning and tool calling, and a 256K context window. This higher-fidelity variant uses ggml-org's Q8_0 llama.cpp build (about 35GB) or mlx-community's matching 8-bit affine conversion (about 33.6GB).",
+  "description": "NVIDIA's Nemotron 3.5 Lightning at 30B parameters — text-only, with native reasoning, native tool calling, and a 256K context window. 8-bit quantization for higher fidelity than the Q4 build, at roughly a third more memory (about 35GB, or 34GB on Apple Silicon).",
   "tags": [
     "nvidia",
     "reasoning",
@@ -22,8 +22,8 @@
     "name": "NVIDIA",
     "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16"
   },
-  "version": "1.0.0",
-  "updatedAt": "2026-08-11T00:00:00Z",
+  "version": "1.0.1",
+  "updatedAt": "2026-08-15T00:00:00Z",
   "minGezelVersion": "1.26223",
   "license": "OpenMDW-1.1",
   "licenseClass": "open",
@@ -129,8 +129,7 @@
     "filename": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Q8_0.gguf",
     "sha256": "8544826247d087745baf4e7540f6911608a61caba7a16702314946af5e4235ae",
     "approxSizeBytes": 35004642304,
-    "quantization": "Q8_0",
-    "residentBytes": 42005570765
+    "quantization": "Q8_0"
   },
   "mlx": {
     "huggingfaceRepo": "mlx-community/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-8bit",
@@ -203,7 +202,6 @@
         "sha256": "ec380f0a6c0c2e294eb2435f5ec521fb2121cfc970e6ebf023d06829714a4eaa",
         "sizeBytes": 430
       }
-    ],
-    "residentBytes": 43650255848
+    ]
   }
 }

package/data/chat-models/ne/nemotron3.5-lightning-30b-q8/versions/1.0.1/manifest.json

@@ -0,0 +1,88 @@
+{
+  "schemaVersion": 1,
+  "version": "1.0.1",
+  "releasedAt": "2026-08-15T00:00:00Z",
+  "approxSizeBytes": 35004642304,
+  "llamaCpp": {
+    "huggingfaceRepo": "ggml-org/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-GGUF",
+    "revision": "88d7ce0b0fa385c5108866ce5d33690927531a37",
+    "filename": "NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Q8_0.gguf",
+    "sha256": "8544826247d087745baf4e7540f6911608a61caba7a16702314946af5e4235ae",
+    "approxSizeBytes": 35004642304,
+    "quantization": "Q8_0"
+  },
+  "mlx": {
+    "huggingfaceRepo": "mlx-community/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-8bit",
+    "revision": "881a85f9e84dac5ab1a61d4493739b5ef67efeaa",
+    "quantization": "8bit",
+    "approxSizeBytes": 33577119883,
+    "files": [
+      {
+        "name": "chat_template.jinja",
+        "sha256": "58933db77d3099b4f78c55a38347a72e1ea05b97d6bd8f38775303dc0194e0a9",
+        "sizeBytes": 9867
+      },
+      {
+        "name": "config.json",
+        "sha256": "a1b0135c0973322d188a836c86746e69ea07c02b241e1f26681bf5123345359b",
+        "sizeBytes": 2932
+      },
+      {
+        "name": "generation_config.json",
+        "sha256": "0ed94823ae737b507aef6af01e837041b7aa2fb0306793a310cd201d89a31ff4",
+        "sizeBytes": 210
+      },
+      {
+        "name": "model-00001-of-00007.safetensors",
+        "sha256": "7ce9d0b7b01bb133ebf089a65dbcce23dcf35d73eb28dfa20b639a21e86711f3",
+        "sizeBytes": 5359226316
+      },
+      {
+        "name": "model-00002-of-00007.safetensors",
+        "sha256": "ad229b96564455b865364db16918a557d55f21199e4ff7a9d0e0952e56c5c9e0",
+        "sizeBytes": 5046688726
+      },
+      {
+        "name": "model-00003-of-00007.safetensors",
+        "sha256": "a2c01a7ae42cc9881768fc3f4c1820a86c00766f8f88f154fd17e32d10ed1a46",
+        "sizeBytes": 4963631598
+      },
+      {
+        "name": "model-00004-of-00007.safetensors",
+        "sha256": "f9afcd3cedfc7e107b7e199fea995d105ee2ad56929fd3dd017ce6dd24ea9f55",
+        "sizeBytes": 5051598289
+      },
+      {
+        "name": "model-00005-of-00007.safetensors",
+        "sha256": "bf8ef73665e3d894d5badd01cef18a99e7da7dcc3a21c52c80a298976499420c",
+        "sizeBytes": 4938766121
+      },
+      {
+        "name": "model-00006-of-00007.safetensors",
+        "sha256": "bbd2d487828c3e2434ca181673023a1489699a6739c2725250ac0c91d5eb1e17",
+        "sizeBytes": 5026732862
+      },
+      {
+        "name": "model-00007-of-00007.safetensors",
+        "sha256": "01b0ace9963a901962e9bbe492d66faa5a0ed47b09ade9da5529375694fa770e",
+        "sizeBytes": 3173319638
+      },
+      {
+        "name": "model.safetensors.index.json",
+        "sha256": "14f9c8d6ff717ea35d64585e4f723544aa0964d7492c10dbf2e27a1ee29a87e5",
+        "sizeBytes": 65410
+      },
+      {
+        "name": "tokenizer.json",
+        "sha256": "623c34567aebb18582765289fbe23d901c62704d6518d71866e0e58db892b5b7",
+        "sizeBytes": 17077484
+      },
+      {
+        "name": "tokenizer_config.json",
+        "sha256": "ec380f0a6c0c2e294eb2435f5ec521fb2121cfc970e6ebf023d06829714a4eaa",
+        "sizeBytes": 430
+      }
+    ]
+  },
+  "minGezelVersion": "1.26227"
+}

package/data/chat-models/or/ornith-35b-q4/manifest.json

@@ -3,7 +3,7 @@
   "kind": "chat-model",
   "id": "ornith-35b-q4",
   "name": "Ornith 1.0 (35B, Q4)",
-  "description": "DeepReinforce's Ornith 1.0 at 35B total parameters (Qwen 3.5 MoE architecture), an agentic-coding model trained with a self-improving reinforcement-learning framework — state-of-the-art on SWE-Bench and Terminal-Bench. Native <think> reasoning, OpenAI-style tool calling, 256K context. Q4_K_M quantization for a ~21GB footprint. MIT licensed.",
+  "description": "DeepReinforce's Ornith 1.0 at 35B parameters — an agentic-coding model built for hands-on software work: reading a codebase, running tools, and seeing a change through. Native reasoning, native tool calling, 256K context. Capable-tier on-device option for machines with 32GB+ memory. MIT licensed.",
   "tags": [
     "deepreinforce",
     "coding",
@@ -18,8 +18,8 @@
     "name": "DeepReinforce",
     "url": "https://huggingface.co/deepreinforce-ai"
   },
-  "version": "1.0.0",
-  "updatedAt": "2026-07-01T00:00:00Z",
+  "version": "1.0.1",
+  "updatedAt": "2026-08-15T00:00:00Z",
   "license": "MIT",
   "licenseClass": "open",
   "licenseShortName": "MIT",
@@ -141,8 +141,7 @@
     "filename": "ornith-1.0-35b-Q4_K_M.gguf",
     "sha256": "ff25291b2599fb927a835e624d2b3540106af61761c3fa57ac4264046dbec002",
     "approxSizeBytes": 21166757760,
-    "quantization": "Q4_K_M",
-    "residentBytes": 25400109312
+    "quantization": "Q4_K_M"
   },
   "mlx": {
     "huggingfaceRepo": "mlx-community/Ornith-1.0-35B-4bit",
@@ -220,7 +219,6 @@
         "sha256": "ce99b4cb2983d118806ce0a8b777a35b093e2000a503ebde25853284c9dfa003",
         "sizeBytes": 6722759
       }
-    ],
-    "residentBytes": 26557921365
+    ]
   }
 }

package/data/chat-models/or/ornith-35b-q4/versions/1.0.1/manifest.json

@@ -0,0 +1,93 @@
+{
+  "schemaVersion": 1,
+  "version": "1.0.1",
+  "releasedAt": "2026-08-15T00:00:00Z",
+  "approxSizeBytes": 21200000000,
+  "llamaCpp": {
+    "huggingfaceRepo": "deepreinforce-ai/Ornith-1.0-35B-GGUF",
+    "revision": "c2e1703039380de4ce6820e97afd185682d3c16c",
+    "filename": "ornith-1.0-35b-Q4_K_M.gguf",
+    "sha256": "ff25291b2599fb927a835e624d2b3540106af61761c3fa57ac4264046dbec002",
+    "approxSizeBytes": 21166757760,
+    "quantization": "Q4_K_M"
+  },
+  "mlx": {
+    "huggingfaceRepo": "mlx-community/Ornith-1.0-35B-4bit",
+    "revision": "781f91090809411b7fc07449817f398a99feb188",
+    "quantization": "4bit",
+    "approxSizeBytes": 20429166971,
+    "files": [
+      {
+        "name": "chat_template.jinja",
+        "sha256": "182e77dd83bd8e9ca818b240b82e28f243762cd5dda32e6eef327df7b1cd107e",
+        "sizeBytes": 7536
+      },
+      {
+        "name": "config.json",
+        "sha256": "550efce0158addbcab424cfc92bdaa297138a949bfc71c724918994b7b692d2f",
+        "sizeBytes": 24151
+      },
+      {
+        "name": "generation_config.json",
+        "sha256": "896eeba60a1195e034dc6b9a7bbddd64b6174992348f5dc7d75daaa93d0bdd45",
+        "sizeBytes": 213
+      },
+      {
+        "name": "model-00001-of-00004.safetensors",
+        "sha256": "2a34b0daa08c1decdeb817301ec4927bfdfa5e573559a3f541f73834383b21d0",
+        "sizeBytes": 5288196062
+      },
+      {
+        "name": "model-00002-of-00004.safetensors",
+        "sha256": "5eab94545fdddd6db8c6d8cd54a3481ba26f2d9fc6d48ee47ea94951c631680b",
+        "sizeBytes": 5368472657
+      },
+      {
+        "name": "model-00003-of-00004.safetensors",
+        "sha256": "8fc467f4036656f488afc86776757ad3cbd4910ab9031f84bb28c55768a904fb",
+        "sizeBytes": 5368324255
+      },
+      {
+        "name": "model-00004-of-00004.safetensors",
+        "sha256": "c700f3b3a89d613dfaf5b430f8e3b70e1aeb0e5b7ad4fe3aebf8c64e3644ce17",
+        "sizeBytes": 4377211327
+      },
+      {
+        "name": "model.safetensors.index.json",
+        "sha256": "0b28df60e33753a14e816d3b31577ae2c93884c58430a4a6de6ae9ea483842ea",
+        "sizeBytes": 215755
+      },
+      {
+        "name": "preprocessor_config.json",
+        "sha256": "27225450ac9c6529872ee1924fcb0962ff5634834f817040f444118116f4e516",
+        "sizeBytes": 390
+      },
+      {
+        "name": "processor_config.json",
+        "sha256": "45fc17c8dd2474af6b493b52483c26c0584b0082d368c480f9fa611e73070040",
+        "sizeBytes": 991
+      },
+      {
+        "name": "tokenizer.json",
+        "sha256": "06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523",
+        "sizeBytes": 19989325
+      },
+      {
+        "name": "tokenizer_config.json",
+        "sha256": "792fa3f0cb88b111e54ef3134c873531008c4df471d108da17903426e308aa7b",
+        "sizeBytes": 1165
+      },
+      {
+        "name": "video_preprocessor_config.json",
+        "sha256": "7768af27c1fafa9cc9011c1dc20067e03f8915e03b63504550e11d5066986d13",
+        "sizeBytes": 385
+      },
+      {
+        "name": "vocab.json",
+        "sha256": "ce99b4cb2983d118806ce0a8b777a35b093e2000a503ebde25853284c9dfa003",
+        "sizeBytes": 6722759
+      }
+    ]
+  },
+  "minGezelVersion": "1.26227"
+}

package/data/chat-models/or/ornith-35b-q8/manifest.json

@@ -3,7 +3,7 @@
   "kind": "chat-model",
   "id": "ornith-35b-q8",
   "name": "Ornith 1.0 (35B, Q8)",
-  "description": "DeepReinforce's Ornith 1.0 at 35B total parameters (Qwen 3.5 MoE architecture), an agentic-coding model trained with a self-improving reinforcement-learning framework — state-of-the-art on SWE-Bench and Terminal-Bench. Native <think> reasoning, OpenAI-style tool calling, 256K context. Q8_0 quantization for higher fidelity than the Q4 build, at roughly double the memory footprint (~37GB). MIT licensed.",
+  "description": "DeepReinforce's Ornith 1.0 at 35B parameters — an agentic-coding model built for hands-on software work: reading a codebase, running tools, and seeing a change through. Native reasoning, native tool calling, 256K context. 8-bit quantization for higher fidelity than the Q4 build, at roughly double the memory footprint (~37GB). MIT licensed.",
   "tags": [
     "deepreinforce",
     "coding",
@@ -18,8 +18,8 @@
     "name": "DeepReinforce",
     "url": "https://huggingface.co/deepreinforce-ai"
   },
-  "version": "1.0.0",
-  "updatedAt": "2026-07-01T00:00:00Z",
+  "version": "1.0.1",
+  "updatedAt": "2026-08-15T00:00:00Z",
   "license": "MIT",
   "licenseClass": "open",
   "licenseShortName": "MIT",
@@ -141,8 +141,7 @@
     "filename": "ornith-1.0-35b-Q8_0.gguf",
     "sha256": "cbc992bca07901c1a51f33e65e6fc5d687de179c852a772dfd15e4c3261dbf5c",
     "approxSizeBytes": 36903138880,
-    "quantization": "Q8_0",
-    "residentBytes": 44283766656
+    "quantization": "Q8_0"
   },
   "mlx": {
     "huggingfaceRepo": "mlx-community/Ornith-1.0-35B-8bit",
@@ -240,7 +239,6 @@
         "sha256": "ce99b4cb2983d118806ce0a8b777a35b093e2000a503ebde25853284c9dfa003",
         "sizeBytes": 6722759
       }
-    ],
-    "residentBytes": 49072879453
+    ]
   }
 }

package/data/chat-models/or/ornith-35b-q8/versions/1.0.1/manifest.json

@@ -0,0 +1,113 @@
+{
+  "schemaVersion": 1,
+  "version": "1.0.1",
+  "releasedAt": "2026-08-15T00:00:00Z",
+  "approxSizeBytes": 37000000000,
+  "llamaCpp": {
+    "huggingfaceRepo": "deepreinforce-ai/Ornith-1.0-35B-GGUF",
+    "revision": "c2e1703039380de4ce6820e97afd185682d3c16c",
+    "filename": "ornith-1.0-35b-Q8_0.gguf",
+    "sha256": "cbc992bca07901c1a51f33e65e6fc5d687de179c852a772dfd15e4c3261dbf5c",
+    "approxSizeBytes": 36903138880,
+    "quantization": "Q8_0"
+  },
+  "mlx": {
+    "huggingfaceRepo": "mlx-community/Ornith-1.0-35B-8bit",
+    "revision": "28d10f45a04981989f5a2e53e38f6473e81e815a",
+    "quantization": "8bit",
+    "approxSizeBytes": 37748365526,
+    "files": [
+      {
+        "name": "chat_template.jinja",
+        "sha256": "182e77dd83bd8e9ca818b240b82e28f243762cd5dda32e6eef327df7b1cd107e",
+        "sizeBytes": 7536
+      },
+      {
+        "name": "config.json",
+        "sha256": "c1cfeef1ed8dc341d400c33d7a2a93b61d01b1a8b70bf8b8992550b09a5dfc07",
+        "sizeBytes": 24151
+      },
+      {
+        "name": "generation_config.json",
+        "sha256": "896eeba60a1195e034dc6b9a7bbddd64b6174992348f5dc7d75daaa93d0bdd45",
+        "sizeBytes": 213
+      },
+      {
+        "name": "model-00001-of-00008.safetensors",
+        "sha256": "356cbcbe9244d120e352782e7f8ab582f8a3125a0a82876c0bc43275b73ad217",
+        "sizeBytes": 5329916013
+      },
+      {
+        "name": "model-00002-of-00008.safetensors",
+        "sha256": "59f16e6cb8bce7a47a5c11458cac47307070c00984a0849e29f1873101f4e102",
+        "sizeBytes": 5365573922
+      },
+      {
+        "name": "model-00003-of-00008.safetensors",
+        "sha256": "449d965928a225fdb5f1d90e4ed0f718a3733dbe03ecc733d65daaf120fca526",
+        "sizeBytes": 5358684668
+      },
+      {
+        "name": "model-00004-of-00008.safetensors",
+        "sha256": "d6754fbc66cbc37c953843aeaf074546f8807860adc1ea7c067b3f747bd09f6c",
+        "sizeBytes": 5365574194
+      },
+      {
+        "name": "model-00005-of-00008.safetensors",
+        "sha256": "d4325c8eb40a2871674ca701340edecc7e753389a3c3ddb11365474a34cf7169",
+        "sizeBytes": 5358684644
+      },
+      {
+        "name": "model-00006-of-00008.safetensors",
+        "sha256": "43a4bd68352772825dfb79c0bd844031818994e953bbd1adb06f3dbb4d9b1e70",
+        "sizeBytes": 5365574174
+      },
+      {
+        "name": "model-00007-of-00008.safetensors",
+        "sha256": "1f6b3423dc7c560ae2ad43781b66ab92fe439cd684ab47be051c4bd45a0032b4",
+        "sizeBytes": 5037050574
+      },
+      {
+        "name": "model-00008-of-00008.safetensors",
+        "sha256": "259f82efaf13be9af40eb427d913a51c8bb340e57bfd8ea2b5cf629a5cb78de6",
+        "sizeBytes": 540344667
+      },
+      {
+        "name": "model.safetensors.index.json",
+        "sha256": "3db12edeebeb65cab9a6eeb63cd74be4e0c74139a75f672701290b98230501cf",
+        "sizeBytes": 215755
+      },
+      {
+        "name": "preprocessor_config.json",
+        "sha256": "27225450ac9c6529872ee1924fcb0962ff5634834f817040f444118116f4e516",
+        "sizeBytes": 390
+      },
+      {
+        "name": "processor_config.json",
+        "sha256": "45fc17c8dd2474af6b493b52483c26c0584b0082d368c480f9fa611e73070040",
+        "sizeBytes": 991
+      },
+      {
+        "name": "tokenizer.json",
+        "sha256": "06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523",
+        "sizeBytes": 19989325
+      },
+      {
+        "name": "tokenizer_config.json",
+        "sha256": "792fa3f0cb88b111e54ef3134c873531008c4df471d108da17903426e308aa7b",
+        "sizeBytes": 1165
+      },
+      {
+        "name": "video_preprocessor_config.json",
+        "sha256": "7768af27c1fafa9cc9011c1dc20067e03f8915e03b63504550e11d5066986d13",
+        "sizeBytes": 385
+      },
+      {
+        "name": "vocab.json",
+        "sha256": "ce99b4cb2983d118806ce0a8b777a35b093e2000a503ebde25853284c9dfa003",
+        "sizeBytes": 6722759
+      }
+    ]
+  },
+  "minGezelVersion": "1.26227"
+}

package/data/chat-models/or/ornith-9b-q4/manifest.json

@@ -3,7 +3,7 @@
   "kind": "chat-model",
   "id": "ornith-9b-q4",
   "name": "Ornith 1.0 (9B)",
-  "description": "DeepReinforce's Ornith 1.0 at 9B parameters (dense, built on Qwen 3.5 + Gemma 4), an agentic-coding model trained with a self-improving reinforcement-learning framework. Native <think> reasoning, OpenAI-style tool calling, 256K context. The small-tier pick for tool-heavy work on a 12GB+ inference budget. Q4_K_M quantization. MIT licensed.",
+  "description": "DeepReinforce's Ornith 1.0 at 9B parameters — an agentic-coding model built for hands-on software work. Native reasoning, native tool calling, 256K context. The small-tier pick for tool-heavy work on a 12GB+ inference budget. MIT licensed.",
   "tags": [
     "deepreinforce",
     "coding",
@@ -17,8 +17,8 @@
     "name": "DeepReinforce",
     "url": "https://huggingface.co/deepreinforce-ai"
   },
-  "version": "1.0.0",
-  "updatedAt": "2026-07-01T00:00:00Z",
+  "version": "1.0.1",
+  "updatedAt": "2026-08-15T00:00:00Z",
   "license": "MIT",
   "licenseClass": "open",
   "licenseShortName": "MIT",
@@ -142,8 +142,7 @@
     "filename": "ornith-1.0-9b-Q4_K_M.gguf",
     "sha256": "5720d1f671b4996481274fffe01868c3c36e87c135cc8538471cc7bd6087b106",
     "approxSizeBytes": 5629108704,
-    "quantization": "Q4_K_M",
-    "residentBytes": 6754930445
+    "quantization": "Q4_K_M"
   },
   "mlx": {
     "huggingfaceRepo": "mlx-community/Ornith-1.0-9B-4bit",
@@ -211,7 +210,6 @@
         "sha256": "ce99b4cb2983d118806ce0a8b777a35b093e2000a503ebde25853284c9dfa003",
         "sizeBytes": 6722759
       }
-    ],
-    "residentBytes": 7770193739
+    ]
   }
 }

package/data/chat-models/or/ornith-9b-q4/versions/1.0.1/manifest.json

@@ -0,0 +1,83 @@
+{
+  "schemaVersion": 1,
+  "version": "1.0.1",
+  "releasedAt": "2026-08-15T00:00:00Z",
+  "approxSizeBytes": 5600000000,
+  "llamaCpp": {
+    "huggingfaceRepo": "deepreinforce-ai/Ornith-1.0-9B-GGUF",
+    "revision": "3296bc7a404871a72ac3f1903f561459c09b5c17",
+    "filename": "ornith-1.0-9b-Q4_K_M.gguf",
+    "sha256": "5720d1f671b4996481274fffe01868c3c36e87c135cc8538471cc7bd6087b106",
+    "approxSizeBytes": 5629108704,
+    "quantization": "Q4_K_M"
+  },
+  "mlx": {
+    "huggingfaceRepo": "mlx-community/Ornith-1.0-9B-4bit",
+    "revision": "1e980b9742a9e554a4d57e90b4c597811fb2fc4e",
+    "quantization": "4bit",
+    "approxSizeBytes": 5977069777,
+    "files": [
+      {
+        "name": "chat_template.jinja",
+        "sha256": "8b4d21a1e70ccbc8849e5e5e6dd32f3acfb2e2060739475909edc6e9b27c7165",
+        "sizeBytes": 7594
+      },
+      {
+        "name": "config.json",
+        "sha256": "e661291d14e6c4e239ae430d7a72ab1c947440ca89c94fc09be71ff0acb45c37",
+        "sizeBytes": 3876
+      },
+      {
+        "name": "generation_config.json",
+        "sha256": "be19d3c48cd249efd6e0327b69916caae8279007dd14c0aad86f7b6b28c5d148",
+        "sizeBytes": 137
+      },
+      {
+        "name": "model-00001-of-00002.safetensors",
+        "sha256": "60e62b00a2ff10d793aed5fe7efc40137e46c9d4028b14c84f1d170eef964beb",
+        "sizeBytes": 5349769710
+      },
+      {
+        "name": "model-00002-of-00002.safetensors",
+        "sha256": "736da49643bb8c64fa081f2415ce624f40e983be1fc590ad30e65af75cdd5a5a",
+        "sizeBytes": 600449850
+      },
+      {
+        "name": "model.safetensors.index.json",
+        "sha256": "4518e4d98baed265cb25c5adb45839d5b69e0b9700f4bd573533392097bbb874",
+        "sizeBytes": 123592
+      },
+      {
+        "name": "preprocessor_config.json",
+        "sha256": "27225450ac9c6529872ee1924fcb0962ff5634834f817040f444118116f4e516",
+        "sizeBytes": 390
+      },
+      {
+        "name": "processor_config.json",
+        "sha256": "45fc17c8dd2474af6b493b52483c26c0584b0082d368c480f9fa611e73070040",
+        "sizeBytes": 991
+      },
+      {
+        "name": "tokenizer.json",
+        "sha256": "06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523",
+        "sizeBytes": 19989325
+      },
+      {
+        "name": "tokenizer_config.json",
+        "sha256": "c908cd42c5e6eac721445fb70841e45845a6a269c6ef2c595e97a578a9724671",
+        "sizeBytes": 1168
+      },
+      {
+        "name": "video_preprocessor_config.json",
+        "sha256": "7768af27c1fafa9cc9011c1dc20067e03f8915e03b63504550e11d5066986d13",
+        "sizeBytes": 385
+      },
+      {
+        "name": "vocab.json",
+        "sha256": "ce99b4cb2983d118806ce0a8b777a35b093e2000a503ebde25853284c9dfa003",
+        "sizeBytes": 6722759
+      }
+    ]
+  },
+  "minGezelVersion": "1.26227"
+}

package/data/chat-models/qw/qwen3.5-122b-a10b-q4/manifest.json

@@ -20,8 +20,8 @@
     "name": "Alibaba",
     "url": "https://huggingface.co/Qwen/Qwen3.5-122B-A10B"
   },
-  "version": "1.0.1",
-  "updatedAt": "2026-07-28T00:00:00Z",
+  "version": "1.0.3",
+  "updatedAt": "2026-08-15T00:00:00Z",
   "license": "Apache-2.0",
   "licenseClass": "open",
   "licenseShortName": "Apache 2.0",
@@ -178,7 +178,6 @@
     ],
     "approxSizeBytes": 78260403200,
     "quantization": "UD-Q4_K_M",
-    "residentBytes": 93912483840,
     "mmproj": {
       "filename": "mmproj-BF16.gguf",
       "sha256": "8f2442758a6f5a2bbfbbddb4b160b08ff4e29c3a76cad7627a0fe2beca2325e6",
@@ -311,7 +310,6 @@
         "sha256": "ce99b4cb2983d118806ce0a8b777a35b093e2000a503ebde25853284c9dfa003",
         "sizeBytes": 6722759
       }
-    ],
-    "residentBytes": 90506802715
+    ]
   }
 }