{ "meta": { "attribution": "Anchor Terminal (https://www.anchorterminal.com)", "attributionRequired": true, "credit": "CC BY 4.0, credit Anchor Terminal and link the page or API address you took it from (https://www.anchorterminal.com/terms/#data)", "docs": "https://www.anchorterminal.com/docs/", "generatedAt": "2026-11-10", "license": "CC-BY-4.0", "method": "https://www.anchorterminal.com/benchmark/", "methodology": "0.7", "openapi": "https://www.anchorterminal.com/openapi.json", "preview": false, "run": "2026-10-01", "runLabel": "October 2026 research run", "terms": "https://www.anchorterminal.com/terms/#data" }, "tool": { "slug": "ornith-1-5-122b", "name": "Ornith-1.5-122B", "vendor": "Ornith", "vendorUrl": "https://ornith.ai", "kind": "weights", "category": "open-weights", "summary": "Ornith-1.5-122B is Ornith's open-weight mixture-of-experts reasoning model for coding and agent work, about 125B parameters on Qwen3.5's architecture with a 262,144-token context, published in BF16, FP8, NVFP4, GGUF and MLX and marked MIT.", "url": "https://www.anchorterminal.com/tools/ornith-1-5-122b", "markdownUrl": "https://www.anchorterminal.com/tools/ornith-1-5-122b.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/ornith-1-5-122b.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/ornith-1-5-122b.json", "dns": "ornith-1-5-122b.grade.anchorterminal.com", "repo": "https://huggingface.co/ornith-ai/Ornith-1.5-122B", "license": "MIT", "transports": [], "packages": [ { "registry": "huggingface", "name": "ornith-ai/Ornith-1.5-122B" }, { "registry": "huggingface", "name": "ornith-ai/Ornith-1.5-122B-GGUF" }, { "registry": "huggingface", "name": "ornith-ai/Ornith-1.5-122B-FP8" }, { "registry": "huggingface", "name": "ornith-ai/Ornith-1.5-122B-NVFP4" }, { "registry": "huggingface", "name": "ornith-ai/Ornith-1.5-122B-MLX-4bit" }, { "registry": "ollama", "name": "ornith-1.5:122b-q4_K_M" } ], "auth": "none", "authNotes": "The weights download from Hugging Face with no account, token or terms to accept. The main repository and Ornith's six build repositories for this size are not gated (checked 2026-11-10).", "pricing": "free", "pricingNotes": "Free to download and use, commercial use included, under the MIT licence named in the card's metadata. The repository has no LICENSE file, and the card's licence link returns 404. Running it yourself costs the hardware, about 250 GB for the BF16 files, 69.7 GB for Ornith's MLX 4-bit build or 75.8 GB for its Q4_K_M GGUF (checked 2026-11-10).", "priceSummary": "Free", "where": "local", "x402": { "level": "no", "evidence": "Weights you download and run yourself, so there is nothing to pay per call and no payment protocol (checked 2026-11-10).", "endpoints": [] }, "toolCount": null, "popularity": { "githubStars": null, "npmWeekly": null, "pypiWeekly": null, "asOf": "2026-11-10" }, "docsUrl": "https://huggingface.co/ornith-ai/Ornith-1.5-122B", "capabilities": [ "model.open-weights" ], "tags": [ "open-weights", "model", "moe", "multimodal", "reasoning", "tool-calling", "coding", "mit", "safetensors", "gguf", "mlx", "fp8", "nvfp4", "self-hosted" ], "lastRelease": "2026-10-24", "graded": true, "anchor": { "graded": true, "score": 61, "grade": "C", "agentReady": false, "rank": 0, "ranked": false, "notRankedWhy": "An open-weight model, graded by the open-weight checklist and ranked among open-weight models, not against tools", "rankOf": 1518, "categoryRank": 59, "methodology": "0.7", "run": "2026-10-01", "scores": { "interface": 85, "reach": 75, "risk": 50, "upkeep": 36 }, "pending": [ "performance", "tasks" ], "breakdown": [ { "key": "capability", "name": "Capability", "weight": 35, "effectiveWeight": 0, "pending": true, "points": 0, "reason": "Pending for this listing. Nobody independent has published results for it in at least two areas yet, so its weight is shared across the other four and the total is capped at 61 until they do or our own runs start." }, { "key": "reach", "name": "Reach", "weight": 25, "effectiveWeight": 38.46, "score": 75, "points": 28.85, "reason": "Memory 10, since Ornith's own MLX 4-bit build is 69.7 GB and needs about 83.6 GB with the fifth added, over 64 GB and inside the reference machine's 107.4 GB, as do its Q4_K_M GGUF, its NVFP4 build and Ollama's build. Runtimes 25, the most the line gives, from vLLM 0.33.0, SGLang 0.5.23, Transformers 5.22.2, llama.cpp's qwen35moe architecture, Ollama's library and LM Studio's catalogue for the qwen35moe architecture. Builds 15, Ornith publishes FP8, NVFP4, GGUF and MLX at several precisions. Terms 15, MIT as named in the card's metadata, with nothing added, read by its terms though the repository has no licence file. Download 10, nothing is gated. Hosted 0, Ornith runs no API and no inference provider serves it. 10 + 25 + 15 + 15 + 10 + 0 = 75.", "checked": "2026-11-10" }, { "key": "risk", "name": "Risk", "weight": 20, "effectiveWeight": 30.77, "score": 50, "points": 15.38, "reason": "File safety 25, safetensors and GGUF only and no Python in any of Ornith's repositories for this size, so the card's --trust-remote-code flag loads nothing from the repository. Provenance 15, the files come from the ornith-ai organisation Ornith's blog links to, with a SHA-256 for every weight file. Safety evaluation 0 and agent guidance 0, nothing published on either. Licence risk 10, MIT carries no policy, threshold or changeable terms, but the card's licence link is broken. Security programme 0, no security.txt, disclosure policy or security contact.", "checked": "2026-11-10" }, { "key": "interface", "name": "Agent interface", "weight": 10, "effectiveWeight": 15.38, "score": 85, "points": 13.08, "reason": "Chat template 20, a Jinja file in the repository. Tool-calling format 25, the card shows how tools go in and how calls come back and names the parser for vLLM and SGLang. Trained tool calling 20, Ornith trains it on agent tasks and the template renders several calls in one turn. Structured output 0, not documented or claimed. Reasoning control 10, Ornith's template reads enable_thinking per request and the card says the reasoning comes back in reasoning_content, though the card doesn't name the switch. Context 10, 262,144 tokens native.", "checked": "2026-11-10" }, { "key": "upkeep", "name": "Documentation and upkeep", "weight": 10, "effectiveWeight": 15.38, "score": 36, "points": 5.54, "reason": "Model card 11, it covers intended use, architecture, parameters and context but not limitations, languages or a knowledge cut-off. Training data 5, a description of the method with no data or cut-off. Last change 20, a folder upload to the MLX builds on 24 October 2026, 17 days ago. Responsiveness 0, Ornith answered none of the 5 questions and reports. Version policy 0, the MLX builds were uploaded a second time four days after the first with no note. File health 0, the card links a missing LICENSE, states hardware its own files don't fit and names a variant that doesn't exist, and the two chat templates differ. 11 + 5 + 20 + 0 + 0 + 0 = 36.", "checked": "2026-11-10" } ], "assessment": { "date": "2026-11-10", "basis": "public evidence", "confidence": "low", "notes": { "capability": "Pending for this listing. Independent results were found in 0 areas (none), and Capability needs 2. Its weight is shared across the other four categories and the total is capped at 61 until there are more, or our own runs start.", "interface": "Chat template 20, a Jinja file in the repository. Tool-calling format 25, the card shows how tools go in and how calls come back and names the parser for vLLM and SGLang. Trained tool calling 20, Ornith trains it on agent tasks and the template renders several calls in one turn. Structured output 0, not documented or claimed. Reasoning control 10, Ornith's template reads enable_thinking per request and the card says the reasoning comes back in reasoning_content, though the card doesn't name the switch. Context 10, 262,144 tokens native.", "reach": "Memory 10, since Ornith's own MLX 4-bit build is 69.7 GB and needs about 83.6 GB with the fifth added, over 64 GB and inside the reference machine's 107.4 GB, as do its Q4_K_M GGUF, its NVFP4 build and Ollama's build. Runtimes 25, the most the line gives, from vLLM 0.33.0, SGLang 0.5.23, Transformers 5.22.2, llama.cpp's qwen35moe architecture, Ollama's library and LM Studio's catalogue for the qwen35moe architecture. Builds 15, Ornith publishes FP8, NVFP4, GGUF and MLX at several precisions. Terms 15, MIT as named in the card's metadata, with nothing added, read by its terms though the repository has no licence file. Download 10, nothing is gated. Hosted 0, Ornith runs no API and no inference provider serves it. 10 + 25 + 15 + 15 + 10 + 0 = 75.", "risk": "File safety 25, safetensors and GGUF only and no Python in any of Ornith's repositories for this size, so the card's --trust-remote-code flag loads nothing from the repository. Provenance 15, the files come from the ornith-ai organisation Ornith's blog links to, with a SHA-256 for every weight file. Safety evaluation 0 and agent guidance 0, nothing published on either. Licence risk 10, MIT carries no policy, threshold or changeable terms, but the card's licence link is broken. Security programme 0, no security.txt, disclosure policy or security contact.", "upkeep": "Model card 11, it covers intended use, architecture, parameters and context but not limitations, languages or a knowledge cut-off. Training data 5, a description of the method with no data or cut-off. Last change 20, a folder upload to the MLX builds on 24 October 2026, 17 days ago. Responsiveness 0, Ornith answered none of the 5 questions and reports. Version policy 0, the MLX builds were uploaded a second time four days after the first with no note. File health 0, the card links a missing LICENSE, states hardware its own files don't fit and names a variant that doesn't exist, and the two chat templates differ. 11 + 5 + 20 + 0 + 0 + 0 = 36." }, "sources": [ { "what": "model card", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B", "seen": "2026-11-10" }, { "what": "gating, dates, parameter count, downloads, file list and hashes", "url": "https://huggingface.co/api/models/ornith-ai/Ornith-1.5-122B?blobs=true", "seen": "2026-11-10" }, { "what": "inference providers, none", "url": "https://huggingface.co/api/models/ornith-ai/Ornith-1.5-122B?expand[]=inferenceProviderMapping", "seen": "2026-11-10" }, { "what": "commit history of the main repository", "url": "https://huggingface.co/api/models/ornith-ai/Ornith-1.5-122B/commits/main", "seen": "2026-11-10" }, { "what": "GGUF files and sizes", "url": "https://huggingface.co/api/models/ornith-ai/Ornith-1.5-122B-GGUF?blobs=true", "seen": "2026-11-10" }, { "what": "GGUF commit history", "url": "https://huggingface.co/api/models/ornith-ai/Ornith-1.5-122B-GGUF/commits/main", "seen": "2026-11-10" }, { "what": "FP8 build sizes", "url": "https://huggingface.co/api/models/ornith-ai/Ornith-1.5-122B-FP8?blobs=true", "seen": "2026-11-10" }, { "what": "NVFP4 build sizes", "url": "https://huggingface.co/api/models/ornith-ai/Ornith-1.5-122B-NVFP4?blobs=true", "seen": "2026-11-10" }, { "what": "MLX 4-bit build sizes, the smallest build", "url": "https://huggingface.co/api/models/ornith-ai/Ornith-1.5-122B-MLX-4bit?blobs=true", "seen": "2026-11-10" }, { "what": "MLX build history, second upload on 24 October 2026", "url": "https://huggingface.co/api/models/ornith-ai/Ornith-1.5-122B-MLX/commits/main", "seen": "2026-11-10" }, { "what": "MLX 4-bit build history", "url": "https://huggingface.co/api/models/ornith-ai/Ornith-1.5-122B-MLX-4bit/commits/main", "seen": "2026-11-10" }, { "what": "architecture, context length and no auto_map", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B/raw/main/config.json", "seen": "2026-11-10" }, { "what": "chat template, tool format and thinking switch", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B/raw/main/chat_template.jinja", "seen": "2026-11-10" }, { "what": "second chat template", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B/raw/main/tokenizer_config.json", "seen": "2026-11-10" }, { "what": "licence link from the card, 404", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B/blob/main/LICENSE", "seen": "2026-11-10" }, { "what": "discussions, replies counted by script from authors, dates and member flags", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B/discussions", "seen": "2026-11-10" }, { "what": "Ornith's organisation and builds for this size", "url": "https://huggingface.co/api/models?author=ornith-ai\u0026search=Ornith-1.5-122B", "seen": "2026-11-10" }, { "what": "base model, licence and parameter count", "url": "https://huggingface.co/api/models/Qwen/Qwen3.5-122B-A10B", "seen": "2026-11-10" }, { "what": "Ornith-1.5 blog post, method and link to the organisation", "url": "https://ornith.ai/ornith_1_5.html", "seen": "2026-11-10" }, { "what": "security.txt, redirects to the home page", "url": "https://ornith.ai/.well-known/security.txt", "seen": "2026-11-10" }, { "what": "Ollama library tags, ornith-1.5:122b-q4_K_M 77 GB and ornith-1.5:122b 73 GB", "url": "https://ollama.com/library/ornith-1.5/tags", "seen": "2026-11-10" }, { "what": "Ollama tag page, Q4_K_M 76 GB, projector 912 MB, MIT text", "url": "https://ollama.com/library/ornith-1.5:122b-q4_K_M", "seen": "2026-11-10" }, { "what": "LM Studio catalogue entry for the qwen35moe architecture", "url": "https://lmstudio.ai/models/qwen/qwen3.5-35b-a3b", "seen": "2026-11-10" }, { "what": "vLLM v0.33.0 registry, Qwen3_5MoeForConditionalGeneration", "url": "https://raw.githubusercontent.com/vllm-project/vllm/v0.33.0/vllm/model_executor/models/registry.py", "seen": "2026-11-10" }, { "what": "SGLang v0.5.23 Qwen3_5MoeForConditionalGeneration", "url": "https://raw.githubusercontent.com/sgl-project/sglang/v0.5.23/python/sglang/srt/models/qwen3_5.py", "seen": "2026-11-10" }, { "what": "Transformers v5.22.2 qwen3_5_moe model", "url": "https://raw.githubusercontent.com/huggingface/transformers/v5.22.2/src/transformers/models/qwen3_5_moe/__init__.py", "seen": "2026-11-10" }, { "what": "llama.cpp b12242 qwen35moe architecture", "url": "https://raw.githubusercontent.com/ggml-org/llama.cpp/b12242/src/llama-arch.cpp", "seen": "2026-11-10" } ], "openQuestions": [ "The card's vLLM command passes --trust-remote-code, which loads nothing here because the repository holds no code. If the flag alone should cost the 10 in file safety, Risk falls to 40", "Reasoning control was given 10 because Ornith's own template has the per-request switch and the card says where the reasoning comes back, though the card doesn't name the switch, the same reading as the other Ornith-1.5 sizes. A reading that needs the card to say it gives 0 and Agent interface 75", "Version policy is 0 from the second folder upload to the MLX builds on 24 October 2026. What changed in it wasn't established, since the files weren't compared across commits. If it only added files, the line is 5 for a first release and upkeep 41", "The 73 GB ornith-1.5:122b tag in Ollama's library is marked for MLX and llama.cpp and carries a modelopt quantisation file. Its precision wasn't established, so it isn't a build row and wasn't used for memory", "No LICENSE file and no notice for Qwen3.5 or Gemma 4 (Apache 2.0). Whether MIT can cover the whole derivative wasn't established", "Only Hugging Face's provider list was checked for hosting. Another provider's own list could carry it and earn hosted 5", "Confidence is low by the checklist's rule, since no independent results were looked up for this model on the day. Artificial Analysis and the leaderboards weren't read", "unchecked: a GitHub organisation or security contact for Ornith, the domain's registration date and the base's licence text", "Time move, last change 20 to 15 after 23 November 2026 (upkeep 36 to 31) unless Ornith's files or cards change" ] }, "negative": 0, "verdict": "The largest Ornith that fits a 128 GB machine, a coding and terminal-agent cousin on Qwen3.5-122B-A10B's architecture with XML tool calls and 262,144 tokens of context. Its 4-bit builds need 84 to 93 GB. The main risk is an absent publisher, with no licence file, safety results or replies, and a card that misstates the hardware.", "bestFor": "Local coding and terminal agents on a 128 GB unified-memory machine or a 96 GB GPU setup, where a clear licence file and an answering publisher aren't required", "strengths": [ "4-bit builds from Ornith itself in GGUF, MLX and NVFP4, at 70 to 78 GB, plus Ollama's Q4_K_M", "Runs in vLLM, SGLang, Transformers, llama.cpp, Ollama and LM Studio from their released versions", "The card names the tool and reasoning parsers for vLLM and SGLang, with a tool-call example", "262,144 tokens of native context", "Safetensors and GGUF only, no gate and no code in the repository" ], "weaknesses": [ "No LICENSE file though the card links one, and no notice for the Apache 2.0 Qwen3.5 and Gemma 4 work it builds on", "No safety evaluation, agent-use guidance or security contact", "The card says a single 80 GB GPU serves it, which its 250 GB of BF16 files don't fit, and names a Mobile variant that isn't published", "Ornith answered none of the 5 discussions, and the MLX builds were uploaded again with no note", "The two chat templates disagree on keeping earlier reasoning, and structured output isn't documented" ], "agentNotes": [ "Start it with `ollama run ornith-1.5:122b-q4_K_M`, or `llama-server -hf ornith-ai/Ornith-1.5-122B-GGUF:Q4_K_M --jinja -c 262144`", "Budget about 92 GB for the Q4_K_M GGUF with a working context, and less context if the machine has under 100 GB for the model", "In vLLM pass `--enable-auto-tool-choice --tool-call-parser qwen3_xml --reasoning-parser qwen3`, and in SGLang `--tool-call-parser qwen3_coder --reasoning-parser qwen3`, so tool calls and reasoning come back parsed", "Thinking is on by default. Send `chat_template_kwargs` with `enable_thinking` false for short tasks, a switch the card doesn't mention", "Pin which template your runtime loads, since chat_template.jinja keeps earlier reasoning and the tokenizer_config.json template drops it", "Use temperature 0.6, top_p 0.95, top_k 20 for coding and temperature 1.0 for general tasks, as the card recommends", "Add mmproj-Ornith-1.5-122B-BF16.gguf (0.9 GB) for image input in llama.cpp, and validate JSON output yourself, since structured output isn't documented" ], "metrics": { "kind": "local", "measured": false }, "reviewCount": 0, "avgRating": 0, "history": [ { "basis": "public evidence", "confidence": "low", "grade": "C", "methodology": "0.7", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 61 } ], "editorialScores": { "interface": 85, "reach": 75, "risk": 50, "upkeep": 36 }, "provenanceScore": 19, "checked": { "categories": { "interface": "2026-11-10", "reach": "2026-11-10", "risk": "2026-11-10", "upkeep": "2026-11-10" }, "last": "2026-11-10", "oldest": "2026-11-10", "stale": false, "staleAfterDays": 90 }, "scoreHistory": [ { "date": "2026-11-10", "grade": "C", "kind": "listed", "method": "0.7", "score": 61 } ] }, "connect": { "install": "ollama run ornith-1.5:122b-q4_K_M\n# or, with llama.cpp and Ornith's own GGUF:\nllama-server -hf ornith-ai/Ornith-1.5-122B-GGUF:Q4_K_M --jinja -c 262144\n# or, with the card's vLLM command (abridged):\nvllm serve ornith-ai/Ornith-1.5-122B --served-model-name Ornith-1.5-122B --max-model-len 262144 \\\n --enable-auto-tool-choice --tool-call-parser qwen3_xml --reasoning-parser qwen3", "http": "curl http://localhost:11434/api/chat \\\n -d '{\"model\": \"ornith-1.5:122b-q4_K_M\", \"messages\": [{\"role\": \"user\", \"content\": \"Hello!\"}]}'" }, "letme": { "capability": "https://letme.dev/model.open-weights", "tool": "https://letme.dev/ornith-1-5-122b" }, "sameCompany": [ "ornith-1-5-35b-a3b", "ornith-1-5-397b", "ornith-1-5-9b" ], "weights": { "trainedBy": "Ornith", "builtOn": "Qwen3.5-122B-A10B, inferred. Not declared in the metadata. Same qwen3_5_moe architecture and an identical parameter count to Qwen3.5-122B-A10B (Apache 2.0); the card says Ornith-1.0 was built on Qwen3.5 and Gemma 4 and that 1.5 extends it", "params": "125B", "context": 262144, "goodFor": "Coding and terminal-agent mixture-of-experts model with tool calling and thinking on by default, text and images in, the largest Ornith that fits a 128 GB machine at 4 bits.", "runtimes": [ "llama.cpp", "Ollama", "vLLM", "SGLang", "Transformers" ], "builds": [ { "name": "Ornith-1.5-122B-NVFP4", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B-NVFP4", "who": "publisher", "precision": "nvfp4", "sizeGB": 77.7, "memoryGB": 96, "format": "safetensors", "note": "NVFP4 needs NVIDIA Blackwell" }, { "name": "Ornith-1.5-122B-Q4_K_M", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B-GGUF/blob/main/Ornith-1.5-122B-Q4_K_M.gguf", "who": "publisher", "precision": "q4_k_m", "sizeGB": 75.8, "memoryGB": 96, "format": "gguf", "runsOn": [ "llama.cpp (Vulkan)", "llama.cpp (ROCm)" ], "note": "Add mmproj-Ornith-1.5-122B-BF16.gguf (0.9 GB) for image input" }, { "name": "Ornith-1.5-122B-MLX-4bit", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B-MLX-4bit", "who": "publisher", "precision": "4bit", "sizeGB": 69.7, "memoryGB": 96, "format": "mlx", "note": "MLX runs on Apple silicon only" }, { "name": "ollama ornith-1.5:122b-q4_K_M", "url": "https://ollama.com/library/ornith-1.5:122b-q4_K_M", "who": "ollama", "precision": "q4_k_m", "sizeGB": 77, "memoryGB": 96, "format": "ollama", "runsOn": [ "Ollama" ], "note": "Q4_K_M (76 GB) with a BF16 vision projector (912 MB), as Ollama's tag page shows it, rounded to whole GB. The 73 GB ornith-1.5:122b tag is marked for MLX and llama.cpp and its quantisation wasn't established, so it isn't listed as a build" }, { "name": "Ornith-1.5-122B-Q6_K", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B-GGUF/blob/main/Ornith-1.5-122B-Q6_K.gguf", "who": "publisher", "precision": "q6_k", "sizeGB": 102.4, "memoryGB": 128, "format": "gguf", "runsOn": [ "llama.cpp (Vulkan)", "llama.cpp (ROCm)" ] }, { "name": "Ornith-1.5-122B-Q5_K_M", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B-GGUF/blob/main/Ornith-1.5-122B-Q5_K_M.gguf", "who": "publisher", "precision": "q5_k_m", "sizeGB": 88.7, "memoryGB": 128, "format": "gguf", "runsOn": [ "llama.cpp (Vulkan)", "llama.cpp (ROCm)" ] }, { "name": "Ornith-1.5-122B-FP8", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B-FP8", "who": "publisher", "precision": "fp8", "sizeGB": 133, "memoryGB": 192, "format": "safetensors", "runsOn": [ "vLLM", "SGLang" ], "note": "Too big for the 128 GB reference machine" }, { "name": "Ornith-1.5-122B-Q8_0", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B-GGUF/blob/main/Ornith-1.5-122B-Q8_0.gguf", "who": "publisher", "precision": "q8_0", "sizeGB": 132.6, "memoryGB": 192, "format": "gguf", "runsOn": [ "llama.cpp (Vulkan)", "llama.cpp (ROCm)" ], "note": "Too big for the 128 GB reference machine" }, { "name": "Ornith-1.5-122B-MLX-8bit", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B-MLX-8bit", "who": "publisher", "precision": "8bit", "sizeGB": 130.7, "memoryGB": 192, "format": "mlx", "note": "MLX runs on Apple silicon only" }, { "name": "Ornith-1.5-122B", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B", "who": "publisher", "precision": "bf16", "sizeGB": 250.2, "memoryGB": 512, "format": "safetensors", "runsOn": [ "vLLM (ROCm)", "SGLang (ROCm)" ], "note": "Too big for the 128 GB reference machine" }, { "name": "Ornith-1.5-122B-BF16", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B-GGUF/blob/main/Ornith-1.5-122B-BF16.gguf", "who": "publisher", "precision": "bf16", "sizeGB": 249.4, "memoryGB": 512, "format": "gguf", "runsOn": [ "llama.cpp (Vulkan)", "llama.cpp (ROCm)" ], "note": "Too big for the 128 GB reference machine" }, { "name": "Ornith-1.5-122B-MLX", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B-MLX", "who": "publisher", "precision": "bf16", "sizeGB": 245.2, "memoryGB": 512, "format": "mlx", "note": "MLX runs on Apple silicon only" } ], "run": "ollama run ornith-1.5:122b-q4_K_M", "runBuild": "ollama ornith-1.5:122b-q4_K_M", "checked": "2026-11-10", "minMemoryGB": 96 }, "family": { "lab": "Ornith", "family": "Ornith-1", "url": "https://www.anchorterminal.com/open-weights/#ornith", "others": [ { "name": "Ornith-1.0-9B", "url": "https://huggingface.co/ornith-ai/Ornith-1.0-9B", "params": "9B (dense)" }, { "name": "Ornith-1.0-35B", "url": "https://huggingface.co/ornith-ai/Ornith-1.0-35B", "params": "35B (MoE)" }, { "name": "Ornith-1.0-397B", "url": "https://huggingface.co/ornith-ai/Ornith-1.0-397B", "params": "397B (MoE)" }, { "name": "Ornith-1.5-9B", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-9B", "params": "9B (dense)", "listing": "https://www.anchorterminal.com/tools/ornith-1-5-9b" }, { "name": "Ornith-1.5-9B-DFlash", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-9B-DFlash", "params": "1.29B (Hub safetensors count)" }, { "name": "Ornith-1.5-35B-A3B", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-35B-A3B", "params": "35B total, 3B active (MoE)", "listing": "https://www.anchorterminal.com/tools/ornith-1-5-35b-a3b" }, { "name": "Ornith-1.5-35B-A3B-DFlash", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-35B-A3B-DFlash", "params": "0.39B (Hub safetensors count)" }, { "name": "Ornith-1.5-397B", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-397B", "params": "397B (MoE)", "listing": "https://www.anchorterminal.com/tools/ornith-1-5-397b" }, { "name": "Ornith-1.5-397B-DFlash", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-397B-DFlash", "params": "1.29B (Hub safetensors count)" } ] }, "notable": [ "262,159 downloads in the last 30 days on Hugging Face, with 219 likes, on 10 November 2026. Ornith's own GGUF repository had 870,881 downloads in 30 days, its FP8 build 413,241 and its NVFP4 build 273,925 (https://huggingface.co/api/models?author=ornith-ai\u0026search=Ornith-1.5-122B)", "The repository was created on 16 October 2026 and the build repositories on 20 October, two months after the 9B, 35B-A3B and 397B sizes of 19 August (https://huggingface.co/api/models/ornith-ai/Ornith-1.5-122B)", "Ollama's library has it as ornith-1.5:122b-q4_K_M, 77 GB (Q4_K_M with a 912 MB BF16 vision projector) with a 262,144-token context set in its parameters, and a second 73 GB tag, ornith-1.5:122b, marked for MLX and llama.cpp (https://ollama.com/library/ornith-1.5/tags)", "The card's table puts it at 84.5 on SWE-bench Verified, 81.2 on Terminal-Bench 2.1 with Claude Code and 74.6 on MCP-Atlas, against 82.8, 83.4 and 74.1 for Qwen3.8-Flash-Next. These are Ornith's numbers with the harnesses and sampling stated, not our measurement (https://huggingface.co/ornith-ai/Ornith-1.5-122B)", "The parameter count, 125,086,497,008 by the Hub's safetensors count, is identical to Qwen3.5-122B-A10B's, and the config is Qwen3_5MoeForConditionalGeneration, so the base is inferred as Qwen3.5-122B-A10B. No base_model is declared in the metadata (https://huggingface.co/api/models/Qwen/Qwen3.5-122B-A10B)", "The card says the model serves on a single 80 GB GPU and names a quantised Ornith-1.5-122B-Mobile variant. The BF16 files are 250 GB, and no repository of that name is in Ornith's organisation (https://huggingface.co/ornith-ai/Ornith-1.5-122B)" ], "area": "models", "details": [ { "label": "Surface graded", "value": "The open weights at huggingface.co/ornith-ai/Ornith-1.5-122B and Ornith's own GGUF, FP8, NVFP4 and MLX repositories, read from the model card, the files and their metadata, the chat templates, Ornith's blog post and the runtimes' own code on 10 November 2026. Nothing was downloaded or run. Ornith runs no hosted API that we found" }, { "label": "Model", "value": "125,086,497,008 parameters in BF16 per the safetensors metadata, the qwen3_5_moe architecture (Qwen3_5MoeForConditionalGeneration), 48 layers in a 3 to 1 mix of linear and full attention, 256 experts with 8 routed per token, a vision encoder and a multi-token prediction layer. Text and images in, text out. The card calls it a 122B mixture-of-experts model and doesn't state the active parameters" }, { "label": "Files", "value": "Sixty-two safetensors shards in BF16 (250.2 GB), a chat_template.jinja and a second chat template in tokenizer_config.json that differs from it, and no Python files. No pickle files. Hugging Face lists a SHA-256 for every shard. No LICENSE file, though the card links one" }, { "label": "Licence", "value": "MIT in the card's metadata, with no licence text, copyright holder or upstream notice in the repository. Ollama's library entry carries an MIT text with the copyright line of deepreinforce.ai. The card says the family builds on Qwen3.5 and Gemma 4, both Apache 2.0, whose notices aren't carried" }, { "label": "Runtimes", "value": "Serving commands for vLLM 0.19.1 or later and SGLang 0.5.9 or later on the card, Transformers 5.8.1 or later, and llama.cpp and Ollama commands for the GGUF. Model code for qwen3_5_moe in vLLM 0.33.0, SGLang 0.5.23 and Transformers 5.22.2, the qwen35moe architecture in llama.cpp b12242, an ornith-1.5:122b-q4_K_M entry in Ollama's library and the qwen35moe architecture in LM Studio's catalogue" }, { "label": "Context", "value": "262,144 tokens native (max_position_embeddings in config.json, rope_type default). The card says YaRN with a factor of 4 reaches about 1,000,000, and warns that static YaRN can cost quality on ordinary inputs" }, { "label": "Tools and reasoning", "value": "The card serves it with `--tool-call-parser qwen3_xml` in vLLM and `qwen3_coder` in SGLang and shows a tool call through the OpenAI-compatible API. The template declares tools in `\u003ctools\u003e` and expects `\u003cfunction=...\u003e` blocks in `\u003ctool_call\u003e`, several per turn. Thinking is on by default and comes back in `reasoning_content`. The template turns it off with `enable_thinking: false`, which the card doesn't mention" }, { "label": "Hardware", "value": "The card says a single 80 GB GPU, which the BF16 files (250 GB) don't fit. Ornith's GGUF files run from 75.8 GB (Q4_K_M) to 249.4 GB (BF16), its MLX 4-bit build is 69.7 GB, NVFP4 77.7 GB and FP8 133.0 GB. With a fifth added, the 4-bit builds need 84 to 93 GB" }, { "label": "Training data", "value": "Ornith's blog describes the method (continued pretraining, mid-training and reinforcement learning on self-generated tasks and scaffolds, on top of Qwen3.5 and Gemma 4) but names no datasets, sizes or knowledge cut-off, and doesn't mention the 122B size" } ], "provenance": { "legalEntity": "", "domain": "ornith.ai", "domainRegistered": "", "endpointOnVendorDomain": null, "terms": "", "privacy": "", "statusPage": "", "changelog": "", "securityTxt": "none", "checked": "2026-11-10", "notes": [ "No legal entity named on the card, the blog or the repository. The citation reads Ornith Team. Ollama's copy of the licence names deepreinforce.ai as the copyright holder.", "Nothing is hosted, so the licence stands in for terms of service and no privacy policy is scored. The licence is MIT in the card's metadata only; the LICENSE file the card links returns 404.", "No changelog for the files. The commit messages are uploads, deletions and README updates.", "ornith.ai/.well-known/security.txt redirects to the home page, so there is no security.txt.", "unchecked: the domain's registration date." ], "score": 19, "checks": [ { "check": "Legal entity named", "value": "not found", "points": 0, "max": 20, "state": "no" }, { "check": "Domain age", "value": "ornith.ai, no registry record we could read", "points": 0, "max": 15, "state": "no" }, { "check": "Endpoint on the vendor's domain", "value": "no hosted endpoint", "points": 0, "max": 0, "state": "na" }, { "check": "Terms of service", "value": "nothing hosted, so the MIT licence stands in", "points": 10, "max": 10, "state": "ok" }, { "check": "Privacy policy", "value": "nothing hosted, not scored", "points": 0, "max": 0, "state": "na" }, { "check": "Status page", "value": "not found", "points": 0, "max": 10, "state": "no" }, { "check": "Changelog", "value": "not found", "points": 0, "max": 10, "state": "no" }, { "check": "security.txt", "value": "not found", "points": 0, "max": 10, "state": "no" }, { "check": "People named", "value": "not researched yet", "points": 0, "max": 0, "state": "na" }, { "check": "Support route", "value": "help pages, a community", "points": 5, "max": 5, "state": "ok" } ] }, "humans": { "slug": "ornith-1-5-122b", "checked": "2026-11-10", "company": "Ornith", "peopleNote": "Copied from the Ornith-1.5-35B-A3B listing (checked 28 October 2026). ornith.ai, its posts on Ornith-1.0 (June 2026), Ornith-1.5 (August 2026) and GrandCode (April 2026), and the Hugging Face model cards are signed by the Ornith Team and name no founder, officer or product lead. The home page is a script-only shell with no text.", "companySocial": { "x": "https://x.com/ornith_" }, "support": { "urls": [ "https://huggingface.co/ornith-ai/Ornith-1.5-122B/discussions" ], "community": [ { "label": "Model community discussions", "url": "https://huggingface.co/ornith-ai/Ornith-1.5-122B/discussions", "kind": "link" } ] }, "sources": [ "https://ornith.ai", "https://ornith.ai/sitemap.xml", "https://ornith.ai/ornith_1_0.html", "https://ornith.ai/cp.html", "https://huggingface.co/ornith-ai/Ornith-1.5-122B" ], "ownership": { "status": "unknown", "checked": "2026-10-28", "sources": [ "https://ornith.ai/ornith_1_0.html", "https://ornith.ai/sitemap.xml", "https://huggingface.co/ornith-ai", "https://officechai.com/ai/deepreinforce-releases-open-source-orinth-1-5-family-of-models-with-solid-benchmarks-and-mit-license/" ] } }, "pageJsonUrl": "https://www.anchorterminal.com/tools/ornith-1-5-122b.json" } }