{"v":1,"id":"model:hf:ornith-ai/Ornith-1.0-9B-GGUF","slug":"model-ornith-ai-ornith-1-0-9b-gguf","kind":"model","category":"llm","title":"Ornith-1.0-9B-GGUF","summary":"Aloha! 🌺 Today, we are releasing Ornith-1.0, a self-improving family of open-source models for agentic coding.","source":{"provider":"hf","ref":"ornith-ai/Ornith-1.0-9B-GGUF","url":"https://huggingface.co/ornith-ai/Ornith-1.0-9B-GGUF","rev":"3296bc7a404871a72ac3f1903f561459c09b5c17","fetchedAt":"2026-10-02T20:52:01.818Z","etag":"W/\"2ab1-/4RyOHQlvvxLIKug0UQ8mTwG2Dk\""},"author":{"name":"ornith-ai","url":"https://huggingface.co/ornith-ai"},"license":{"spdx":"mit","raw":"mit","url":"https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B-GGUF/blob/main/LICENSE","open":true},"metrics":{"downloads":2570756,"downloadsWeek":327838,"likes":673,"stars":130156,"openIssues":2528,"lastRelease":{"tag":"v0.5.0","at":"2026-09-23T20:50:06Z"},"pushedAt":"2026-10-02T20:17:08Z","takenAt":"2026-10-02T20:52:01.818Z"},"tags":["transformers","gguf","text-generation","endpoints_compatible","conversational"],"pipeline":"text-generation","links":{"github":"ggml-org/llama.cpp","npm":"node-llama-cpp"},"updatedAt":"2026-06-25T14:12:12.000Z","collectedAt":"2026-10-02T20:52:01.818Z","review":{"numbers":["2,570,756 downloads on Hugging Face","673 likes","license mit","5.2 GB for ornith-1.0-9b-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":"2026-10-02T20:52:01.818Z","http":200},"description":"[](https://deep-reinforce.com/ornith.html)\n\n# Ornith-1.0-9B-GGUF\n\nAloha! 🌺 Today, we are releasing Ornith-1.0, a self-improving family of open-source models for agentic coding. \n\nHighlights: \n\n- **State-of-the-Art Coding Agents**: Available in 9B-Dense, 31B-Dense, 35B-MoE, and 397B-MoE (post-trained on top of Gemma 4 and Qwen 3.5), achieving state-of-the-art performance among open-source models of comparable size on coding benchmarks such as Terminal-Bench 2.1, SWE-Bench, NL2Repo and OpenClaw. \n- **Self-Improving Training Framework**:  Ornith-1.0 employs RL to learn to generate not only solution rollouts, but also the scallfold that drive those rollouts. By jointly optimizing the scaffold and the resulting solution, the model  discovers better search trajectories and generates higher-quality solutions. \n- **Licence**: MIT licensed, globally accessible, and free from regional limitations.\n\n## Ornith 1.0 9B \n\nThis model card documents **Ornith-1.0-9B**, the most lightweight member of the Ornith family, designed for efficient single-GPU deployment.\n\n### Benchmarks\n\nOrnith-1.0-9B\nQwen3.5-9B\nQwen3.5-35B\nGemma4-12B\nGemma4-31B\n\nAgentic Coding\n\nTerminal-Bench 2.1 (Terminus-2)\n43.1\n21.3\n41.4\n21\n42.1\n\nTerminal-Bench 2.1 (Claude Code)\n40.6\n18.9\n38.9\n-\n-\n\nSWE-bench Verified\n69.4\n53.2\n70\n44.2\n52\n\nSWE-bench Pro\n42.9\n31.3\n44.6\n27.6\n35.7\n\nSWE-bench Multilingual\n52\n39.7\n60.3\n32.5\n51.7\n\nNL2Repo\n27.2\n16.2\n20.5\n10.3\n15.5\n\nClaw-eval Avg\n63.1\n53.2\n65.4\n32.5\n48.5\n\nSWE Atlas - QnA\n17.9\n9.2\n13.2\n-\n-\n\nSWE Atlas - RF\n16.6\n4.3\n10.2\n-\n-\n\nSWE Atlas - TW\n15.3\n4.4\n9.8\n-\n-\n\n* Terminal-Bench 2.1 (Terminus-2): We evaluate Terminal-Bench 2.1 using the Harbor/Terminus-2 framework with parser=json, temperature=1.0, top_p=1.0, and a 128K context window. Each run uses a 4-hour timeout with 32 CPU cores and 48GB RAM, and results are averaged over 5 runs. We adjust the Qwen chat template to ensure consistency between training and inference (http…\n\nSource: https://huggingface.co/ornith-ai/Ornith-1.0-9B-GGUF","install":{"kind":"model","hfId":"ornith-ai/Ornith-1.0-9B-GGUF","gated":false,"format":"gguf","files":[{"name":"ornith-1.0-9b-Q4_K_M.gguf","size":5629108704,"quant":"Q4_K_M","sha256":"5720d1f671b4996481274fffe01868c3c36e87c135cc8538471cc7bd6087b106"},{"name":"ornith-1.0-9b-Q5_K_M.gguf","size":6467969472,"quant":"Q5_K_M","sha256":"d1b36095636c096b04ea09e798a7a378956f2fa9099340bd54add1954aaf149c"},{"name":"ornith-1.0-9b-Q6_K.gguf","size":7359259072,"quant":"Q6_K","sha256":"33b6f6a3e3f05078438e12df8a4b55c8acf78ceadcc639d2af1cf35a026e8387"},{"name":"ornith-1.0-9b-Q8_0.gguf","size":9527500992,"quant":"Q8_0","sha256":"d0e4bebaa8b3450c62090df1408f2ee5ccb2094f9c610ffde564a654483d4f37"},{"name":"ornith-1.0-9b-bf16.gguf","size":17920696512,"quant":"BF16","sha256":"27bc753487eed85539c3aef63dd602b79cd060401b928c9ff7d30d5556eca260"}],"totalBytes":46904534752,"suggestedFile":"ornith-1.0-9b-Q4_K_M.gguf","requirements":{"ramGb":7,"diskBytes":5629108704,"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-1.0-9B-GGUF"}}