{"v":1,"id":"model:hf:ornith-ai/Ornith-1.5-9B-GGUF","slug":"model-ornith-ai-ornith-1-5-9b-gguf","kind":"model","category":"llm","title":"Ornith-1.5-9B-GGUF","summary":"Chirp Chirp! 🐦 We are introducing Ornith-1.5, a major step toward building foundation models through end-to-end self-improvement.","source":{"provider":"hf","ref":"ornith-ai/Ornith-1.5-9B-GGUF","url":"https://huggingface.co/ornith-ai/Ornith-1.5-9B-GGUF","rev":"abdd624b12ebf020b767fff532ff44fe552b28c3","fetchedAt":"2026-10-02T20:54:16.962Z","etag":"W/\"2c5d-ypsIyur5XI/vDN2J75KL3T7bicM\""},"author":{"name":"ornith-ai","url":"https://huggingface.co/ornith-ai"},"license":{"spdx":"mit","raw":"mit","url":"https://huggingface.co/ornith-ai/Ornith-1.5-9B/blob/main/LICENSE","open":true},"metrics":{"downloads":5051736,"downloadsWeek":327838,"likes":479,"stars":130156,"openIssues":2528,"lastRelease":{"tag":"v0.5.0","at":"2026-09-23T20:50:06Z"},"pushedAt":"2026-10-02T20:17:08Z","takenAt":"2026-10-02T20:54:16.962Z"},"tags":["transformers","gguf","text-generation","endpoints_compatible","conversational"],"pipeline":"text-generation","links":{"github":"ggml-org/llama.cpp","npm":"node-llama-cpp"},"updatedAt":"2026-08-24T02:45:24.000Z","collectedAt":"2026-10-02T20:54:16.962Z","review":{"numbers":["5,051,736 downloads on Hugging Face","479 likes","license mit","5.4 GB for Ornith-1.5-9B-Q4_K_M.gguf","130,156 stars on ggml-org/llama.cpp","2,528 open issues and PRs","last release v0.5.0 on 2026-09-23","327,838 npm downloads a week for node-llama-cpp","latest node-llama-cpp@3.22.1"],"log":null},"trust":"unlabeled","health":{"status":"alive","checkedAt":"2026-10-02T20:54:16.962Z","http":200},"description":"# Ornith-1.5-9B\n\nChirp Chirp! 🐦 We are introducing Ornith-1.5, a major step toward building foundation models through end-to-end self-improvement.\n\nOrnith-1.5 extends Ornith-1.0 (which was developed on top of Qwen3.5 and Gemma4 with additional continued pretraining, mid-training, and post-training) by expanding the self-improvement loop from scaffold and rollout optimization to jointly optimizing task generation, scaffold construction, and solution rollouts. Rather than relying on a fixed set of human-curated tasks and manually designed harnesses, Ornith-1.5 continuously generates new training tasks, discovers effective strategies for solving them, and improves the policy through reinforcement learning. For more details on the task, harness, and rollout reward design, please refer to our [blog](https://ornith.ai/ornith_1_5.html).\n\n## Ornith 1.5 9B\n\nThis model card documents **Ornith-1.5-9B**, the most lightweight member of the Ornith-1.5 family — a 9B dense model designed for efficient single-GPU deployment, and edge-deployable on mobile devices via its quantized Ornith-1.5-9B-Mobile variant.\n\n### Benchmarks\n\nOrnith-1.5-9B\nOrnith-1.0-9B\nQwen3.5-9B\nQwen3.6-35B-A3B\nGemma-4-31B \n\nCoding\n\nTerminal-Bench 2.1 (Terminus-2)\n46.2\n43.1\n21.3\n52.5\n42.1\n\nTerminal-Bench 2.1 (Claude Code)\n47\n40.6\n18.9\n49.2\n-\n\nSWE-bench Verified\n70.6\n69.4\n53.2\n73.4\n52\n\nSWE-bench Pro\n47.5\n42.9\n31.3\n49.5\n35.7\n\nSWE-bench Multilingual\n54.4\n52\n39.7\n67.2\n51.7\n\nNL2Repo\n32.4\n27.2\n16.2\n29.4\n15.5\n\nSWE Atlas - QnA\n20.6\n17.9\n9.2\n15.5\n-\n\nReasoning\n\nHLE (no tools)\n20.2\n16.8\n14.7\n21.4\n19.5\n\nHLE (with tools)\n30.5\n26.4\n24.5\n28.9\n26.5\n\nGPQA Diamond\n86.4\n82.5\n81.7\n86\n84.3\n\nAgentic\n\nMCP-Atlas\n54.2\n49.4\n46.8\n62.8\n55\n\nToolathlon-Verified\n41.2\n33.4\n29.6\n41.7\n52.8\n\nWideSearch\n59.5\n55.8\n53.6\n60.1\n54.2\n\nBrowseComp\n56.4\n44.8\n41.5\n62\n-\n\nClawEval\n66.5\n63.1\n53.2\n68.7\n48.5\n\n* All results reported for Ornith-1.5 are averaged over five independent runs.\n* Terminal-Benc…\n\nSource: https://huggingface.co/ornith-ai/Ornith-1.5-9B-GGUF","install":{"kind":"model","hfId":"ornith-ai/Ornith-1.5-9B-GGUF","gated":false,"format":"gguf","files":[{"name":"Ornith-1.5-9B-BF16.gguf","size":18407321184,"quant":"BF16","sha256":"a72af9b7f34727a49e030627092e7ebd782e9c3459a0c95afa4b9a07a8650f50"},{"name":"Ornith-1.5-9B-Q4_K_M.gguf","size":5780090816,"quant":"Q4_K_M","sha256":"70c112196e0b7023803c9762752e46d29e612a92c83f995bc3ba1ceb07e8fab6"},{"name":"Ornith-1.5-9B-Q5_K_M.gguf","size":6642544576,"quant":"Q5_K_M","sha256":"e4d9634a3b6546a5c00a8680568fe1125f6c98c704ee51ae52ba07650fb4247d"},{"name":"Ornith-1.5-9B-Q6_K.gguf","size":7558901696,"quant":"Q6_K","sha256":"b6f76e74f86245b3caee014b797c10dca931c4dfdaabfb134eab655f81e4154a"},{"name":"Ornith-1.5-9B-Q8_0.gguf","size":9786060384,"quant":"Q8_0","sha256":"22086870b009dbe9815ee752c48a82de930118a7c5ce5599590892ae03b8b010"},{"name":"mmproj-Ornith-1.5-9B-BF16.gguf","size":921704672,"quant":"BF16","sha256":"626f9f90627402a6bf4a999111d0fbd69b5fcca7aa8ba089d69e5f10e8858e1d"}],"totalBytes":49096623328,"suggestedFile":"Ornith-1.5-9B-Q4_K_M.gguf","requirements":{"ramGb":7,"diskBytes":5780090816,"note":"estimate: suggested file size × 1.15 + 0.5 GB; a real measurement comes with lsh models"},"runWith":["llama.cpp","ollama"],"command":"lsh models install hf:ornith-ai/Ornith-1.5-9B-GGUF"}}