Skip to main content

Overview

Same architecture as Qwen 3.5 9B: 8 full-attention layers, 256K positions, tool calling. LoRA uses the 128K window measured on that base.

Ornith 1.5 9B trains on your agent's own traces, can be benchmarked against your production model before any traffic moves, and serves through the same OpenAI-compatible API as every Overmind model. The final weights are yours to download and run anywhere.

Good for

  • Agents that start from the Ornith 1.5 9B checkpoint
  • LoRA on a 9B Qwen 3.5 architecture
  • Tool-calling rows

More in Ornith

Specs & cost

Cost to trainfrom $0.61
Cost to run (1M tok)from $6.00
GroupOrnith
FamilyOrnith
Parameters9B
TierSmall
Context window256K
Max training context128K
Tool callingYes
MethodsLoRA

On Overmind

  1. 01 Train on your data

    Fine-tune on a training dataset from your agent's traces. Overmind recommends tiers from your data stats; you can edit every setting.

  2. 02 Benchmark vs production

    Each finished model is scored on your eval set with your rubric against the model your agent runs today. The headline is the baseline delta per metric.

  3. 03 Serve and own the weights

    Succeeded runs deploy automatically to Inference through the same OpenAI-compatible API. The final weights are yours to download and run anywhere.