# Ornith

37 Hacker News commenters recommended Ornith and 19 criticized it, November 2025 to November 2026. It was recommended in 5 of 13 months.

**Verdict:** Local models that people find fast and useful for coding, often ahead of earlier Qwen releases. Critics say coding results are uneven and benchmark claims are unproven.

## Why people recommend it

- Good at coding tasks (4 comments): "Based on my limited usage, it is good, gives creative solutions to coding problems." ([source](https://news.ycombinator.com/item?id=48724416)) "which has performed slightly better and faster at Python, db and some backend tasks" ([source](https://news.ycombinator.com/item?id=49155200)) "it's usable for real coding and sys admin work, and works great in opencode." ([source](https://news.ycombinator.com/item?id=50097748))
- Beats earlier Qwen models (3 comments): "it's definitely better than Qwen3.6 35b-a3b which was my go-to before" ([source](https://news.ycombinator.com/item?id=49481526)) "I found Ornith1.5 to be a better MoE model for me, also very fast" ([source](https://news.ycombinator.com/item?id=49750186)) "It's been a pretty decent step up for me compared to Qwen3.6" ([source](https://news.ycombinator.com/item?id=49436238))
- Fast on local hardware (4 comments): "playing with ornith-35B with a draft vibethinker-3b as a draft gave me some good speed/results" ([source](https://news.ycombinator.com/item?id=48723380)) "I found Ornith1.5 to be a better MoE model for me, also very fast" ([source](https://news.ycombinator.com/item?id=49750186)) "much faster on my hardware but better responses" ([source](https://news.ycombinator.com/item?id=49955899))
- Useful for everyday tasks (4 comments): "it's pretty nice to just let it work away at simpler prompts overnight or during lunch break" ([source](https://news.ycombinator.com/item?id=49490996)) "Ornith 1.0 had problems with world knowledge" ([source](https://news.ycombinator.com/item?id=49487694)) "so far its really impressive for my needs" ([source](https://news.ycombinator.com/item?id=49370746))

## What people warn about

- Struggles on some coding tasks (5 comments): "it reached its limits totally botching a longer (simple) kernel implementation job" ([source](https://news.ycombinator.com/item?id=48732119)) "It got rather tangled up when I tried it with one of my coding tests" ([source](https://news.ycombinator.com/item?id=48723340)) "the Qwen-based 35b MoE Ornith version performs worse than Qwen 3.6 and Qwen AgentWorld on my benchmarks" ([source](https://news.ycombinator.com/item?id=48726571))
- Benchmark claims questioned (4 comments): "they just put graphs with different color prioritizing themselves" ([source](https://news.ycombinator.com/item?id=48725336)) "These are simply benchmaxxed versions of either Qwen or Gemma 4." ([source](https://news.ycombinator.com/item?id=48724280)) "there are no benchmarks or weights for it anywhere" ([source](https://news.ycombinator.com/item?id=48724527))
- Loops and hallucinates in tool calls (2 comments): "Long session tool calls sucks and hallucinate a lot with that too." ([source](https://news.ycombinator.com/item?id=48725316)) "seems to get in the loop often specially with tool calls." ([source](https://news.ycombinator.com/item?id=49532370))
- Worse than Qwen 3.6 in tests (2 comments): "the Qwen-based 35b MoE Ornith version performs worse than Qwen 3.6 and Qwen AgentWorld on my benchmarks" ([source](https://news.ycombinator.com/item?id=48726571)) "it performed worse than Qwen3.6-27b in my basic test" ([source](https://news.ycombinator.com/item?id=48725561))

## Alternatives hackers recommend

- [Qwen 3.8 27B](https://hackerapproved.com/p/qwen-3-8-27b.md): 1173 recommend
- [Z.ai GLM](https://hackerapproved.com/p/glm-5-2.md): 1043 recommend
- [Kimi](https://hackerapproved.com/p/kimi-k3.md): 931 recommend
- [Gemma 4](https://hackerapproved.com/p/gemma-4.md): 713 recommend
- [Qwen](https://hackerapproved.com/p/qwen.md): 474 recommend

## Cite as

Hacker Approved, "Ornith", https://hackerapproved.com/p/ornith-1-0/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).