GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
kavirubc
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
kavirubc
Model Tree
Qwen/Qwen3-8B
Input Modalities
Output Modalities
Supported Functionality
Given a concurrent Go program and a partial execution trace (goroutine scheduler events), predict the next scheduler event:
markdown
Input: Go source + partial trace (GoStart, GoBlock, GoUnblock, GoCreate, GoEnd, GoSched)Output: {"event_type": "GoBlock", "goroutine_id": 3, "reasoning": "...", "confidence": "high"}
| Setting | Value |
|---|---|
| Base model | Qwen/Qwen3-8B |
| Method | Unsloth + QLoRA, stratified trajectory-level |
| Dataset | kavirubc/weave-bench (data/train_point_dups_balanced.jsonl) |
| Train examples | 2,004 (stratified/balanced across event classes) |
| Epochs | 3 |
| train_loss | 0.0255 |
python
from transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModelimport torchbase = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-8B", torch_dtype=torch.float16, device_map="auto")model = PeftModel.from_pretrained(base, "kavirubc/weave-ccwm-qwen3-8b-stratified-lora")tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-8B")
bibtex
@misc{weave2026,author = {Hapuarachchi, Kaviru},title = {Weave: Concurrent Code World Models},year = {2026},url = {https://arxiv.org/abs/2606.17508}}