GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
ciocan
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
ciocan
Model Tree
ornith-ai/Ornith-1.5-9B
Input Modalities
Output Modalities
Supported Functionality
282 multi-turn tool-call trajectories (system prompt, task <id> user turn, assistant tool calls, tool outputs truncated to 6000 characters, ending at the submit call) produced by the same agent harness driven by DeepSeek v4 flash on the SpreadsheetBench verified-400 set, keeping only tasks the evaluator scored as passing. 58 failing tasks and 60 passing tasks were held out. Golden workbooks were never in the training text; they only selected which trajectories to keep.
LoRA on all linear projections, rank 32, alpha 64, dropout 0.05; bf16; max sequence length 32768, no packing; 2 epochs; learning rate 1e-4 cosine; batch 1 with 8-step gradient accumulation (72 optimizer steps); gradient checkpointing; loss on assistant tokens only via a custom collator (the chat template has no generation tags). TRL 1.12, PEFT 0.20, transformers 5.16, torch 2.14, one H100, 97 minutes. Train loss 0.58 to 0.55.
Merge into the base with PEFT merge_and_unload() and serve with vLLM 0.28: --tool-call-parser qwen3_xml --reasoning-parser qwen3 --enable-auto-tool-choice --max-model-len 32768, thinking off through chat_template_kwargs: {"enable_thinking": false}, temperature 0.
Held out (118 tasks): base 9B 45, this adapter 29, DeepSeek-driven harness 82. Full 400: 0.34 pass rate (the base 9B run of the same 400 at higher concurrency scored 0.19). The adapter makes the model quit the tool loop earlier than the base; the repo README explains the likely causes and the next round.