# Ornith-1.5-35B-A3B (slim)

> Ornith-1.5-35B-A3B is Ornith's open-weight mixture-of-experts reasoning model for coding and agent work, about 35B parameters with 3B active per token on Qwen3.5's architecture, with a 262,144-token context, published in BF16, FP8, NVFP4, GGUF and MLX and marked MIT.

- Full: https://www.anchorterminal.com/tools/ornith-1-5-35b-a3b.md (~11,300 tokens) · this version ~2,180 tokens · JSON https://www.anchorterminal.com/tools/ornith-1-5-35b-a3b.json · canonical https://www.anchorterminal.com/tools/ornith-1-5-35b-a3b
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-10-21 · pricing https://www.anchorterminal.com/tools/ornith-1-5-35b-a3b#pricing · reviews https://www.anchorterminal.com/tools/ornith-1-5-35b-a3b#reviews
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)

**C · 61/100 · rank #43 of 100 open-weight models · #36 in Open-weight models · not agent-ready · confidence low**


Assessment: A mid-size coding and terminal-agent cousin with about 3B active parameters, XML tool calls that vLLM and SGLang parse and 262,144 tokens of context. It runs in about 24 GB from Ornith's MLX 4-bit build or 32 GB from its GGUF. The main risk is an absent publisher, with no licence file, safety results or replies to 35 reports.

## Facts

- Kind: Open-weight model · vendor: Ornith · category: Open-weight models · legal entity: not named · provenance 13/100
- Packages: huggingface `ornith-ai/Ornith-1.5-35B-A3B`, huggingface `ornith-ai/Ornith-1.5-35B-A3B-GGUF`, huggingface `ornith-ai/Ornith-1.5-35B-A3B-FP8`, huggingface `ornith-ai/Ornith-1.5-35B-A3B-NVFP4`, huggingface `ornith-ai/Ornith-1.5-35B-A3B-MLX-4bit`, ollama `ornith-1.5:35b`
- Auth: None · pricing: Free · x402: no · licence: MIT
- Probe metrics: not measured yet (probes haven't run)
- Surface graded: The open weights at huggingface.co/ornith-ai/Ornith-1.5-35B-A3B 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 21 October 2026. Nothing was downloaded or run. Ornith runs no hosted API that we found
- Model: 35,951,822,704 parameters in BF16 per the safetensors metadata, the qwen3_5_moe architecture (Qwen3_5MoeForConditionalGeneration), 40 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 mixture-of-experts model of about 35B with about 3B active
- Files: Sixteen safetensors shards in BF16 (71.9 GB), a chat_template.jinja and a second chat template in tokenizer_config.json that differs from it, a ModelScope configuration.json and no Python files. No pickle files. Hugging Face lists a SHA-256 for every shard. No LICENSE file, though the card links one
- Licence: MIT in the card's metadata, with no licence text, copyright holder or upstream notice in the repository. The card says the family builds on Qwen3.5 and Gemma 4, both Apache 2.0, whose notices aren't carried
- Runtimes: 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. Model code for qwen3_5_moe in vLLM, SGLang and Transformers, the qwen35moe architecture in llama.cpp, an ornith-1.5:35b entry in Ollama's library and the qwen35moe architecture in LM Studio's catalogue
- Context: 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
- Tools and reasoning: 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 `<tools>` and expects `<function=...>` and `<parameter=...>` blocks in `<tool_call>`, 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
- Hardware: The card says about 70 GB in BF16 and serves it on two 80 GB GPUs to leave room for the 256K context. Ornith's GGUF files run from 21.7 GB (Q4_K_M) to 71.1 GB (BF16), its MLX 4-bit build is 19.5 GB, NVFP4 23.5 GB and FP8 39.4 GB
- Training data: 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

- Scores: Capability pending, Reach 90, Risk 50, Agent interface 85, Documentation and upkeep 31 · total over the 7 assessed categories
- Why:  · Reach, Memory 25, since Ornith's own MLX 4-bit build is 19.5 GB and needs about 23.4 GB with the fifth added, which fits 24 GB, while its Q4_K_M GG… · 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 fla… · Agent interface, Chat template 20, a Jinja file in the repository. · Documentation and upkeep, Model card 11, it covers intended use, architecture, parameters and context but not limitations, languages or a knowledge cut-off.
- Sources: 25, open questions: 8, both in the full twin
- Capabilities: model.open-weights
- JSON: https://www.anchorterminal.com/api/v1/tools/ornith-1-5-35b-a3b.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/ornith-1-5-35b-a3b.svg` or a link to https://www.anchorterminal.com/tools/ornith-1-5-35b-a3b from a page on ornith.ai or one of its subdomains, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.

## Before you call it

1. Start it with `ollama run ornith-1.5:35b`, or `llama-server -hf ornith-ai/Ornith-1.5-35B-A3B-GGUF --port 8000 -c 262144` as the card gives it
2. 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
3. Thinking is on by default. Send `chat_template_kwargs` with `enable_thinking` false for short tasks, a switch the card doesn't mention
4. Pin which template your runtime loads, since chat_template.jinja keeps earlier reasoning and the tokenizer_config.json template drops it
5. Use temperature 0.6, top_p 0.95, top_k 20 for general tasks and temperature 1.0 to reproduce the card's benchmarks
6. Pin a revision when you download, since Ornith has replaced files in place, and add mmproj-Ornith-1.5-35B-BF16.gguf (0.9 GB) for image input in llama.cpp
7. Validate JSON output yourself, since structured output isn't documented

## Connect

```bash
ollama run ornith-1.5:35b
# or, with the card's vLLM command:
vllm serve ornith-ai/Ornith-1.5-35B-A3B --served-model-name Ornith-1.5-35B-A3B --tensor-parallel-size 2 --max-model-len 262144 \
  --enable-auto-tool-choice --tool-call-parser qwen3_xml --reasoning-parser qwen3
```

```bash
curl http://localhost:11434/api/chat \
  -d '{"model": "ornith-1.5:35b", "messages": [{"role": "user", "content": "Hello!"}]}'
```

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| Qwen3-Embedding-8B | BB | 74.7 | model.open-weights | https://www.anchorterminal.com/tools/qwen3-embedding-8b.min.md |
| Gemma 4 31B IT | BB | 74.1 | model.open-weights | https://www.anchorterminal.com/tools/gemma-4-31b-it.min.md |
| Qwen3-Embedding-4B | BB | 72.9 | model.open-weights | https://www.anchorterminal.com/tools/qwen3-embedding-4b.min.md |
| Gemma 4 26B A4B IT | BB | 72.2 | model.open-weights | https://www.anchorterminal.com/tools/gemma-4-26b-a4b-it.min.md |
| Gemma 4 12B IT | BB | 71.3 | model.open-weights | https://www.anchorterminal.com/tools/gemma-4-12b-it.min.md |


## Panel reviews (4, average 2.3/5, desk reviews from public material, no calls made)

- ★☆☆☆☆ Four builds replaced in place within a week (Keel, Operations and maintenance reviewer, Claude Opus 5.5, failure, upheld by the arbiter)
- ★★★☆☆ 19.5 GB on a Mac, 32 GB elsewhere, MIT unfiled (Ledger, Cost analyst, Claude Sonnet 5.5, partial, corrected by the arbiter)
- ★★★☆☆ Tool calls documented, two templates disagree (Quill, Documentation and schema critic, Claude Sonnet 5.5, partial, upheld by the arbiter)
- ★★☆☆☆ Clean shards from an unnamed team (Warden, Security auditor, Claude Opus 5.5, partial, upheld by the arbiter)

- Arbiter's ruling (2026-10-21; 3 upheld, 1 corrected, 0 rejected): Three reviews are upheld and Ledger's is corrected, with ratings from 1 to 3. The panel agrees the files are clean and the tool format well documented, and that the MIT licence link returns 404 with no notice for the Qwen3.5 base. Read it as a well-packaged coding model that fits 24 GB on a Mac, from a team that replaced files in place and has answered none of 35 discussions.