GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
ktwu01
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
ktwu01
Model Tree
Qwen/Qwen3.5-9B
Input Modalities
Output Modalities
Supported Functionality
QLoRA (nf4 base, bf16 compute), r=32, alpha=64, dropout 0, targets: full-attention q/k/v/o_proj, Gated DeltaNet in_proj_qkv/in_proj_z/out_proj, MLP gate/up/down_proj (language model only), lr 1e-4 cosine, 3% warmup, global batch 16 (union: 15), 16,384-token context, 1 epoch, assistant-only loss, seed 20261018.
Decontaminated with 13-gram overlap against all four benchmarks; see data/README.md in the code repo.
python
from transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModelbase = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-9B", revision="c202236235762e1c871ad0ccb60c8ee5ba337b9a", dtype="bfloat16")model = PeftModel.from_pretrained(base, "ktwu01/qwen35-9b-takehome-expert-sci") # pin revision=... from the candidate repo READMEtok = AutoTokenizer.from_pretrained("Qwen/Qwen3.5-9B", revision="c202236235762e1c871ad0ccb60c8ee5ba337b9a")See the candidate repository README for results, confidence intervals and protocols.
Apache-2.0 (as the base model). Training data licenses above; OpenScienceReasoning-2 is CC-BY-4.0 and requires attribution to NVIDIA. Research artifact: evaluated under the experiment-specific protocol in the code repository (apptainer environment, FP8 serving); scores are not official leaderboard results.