GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
AnkitAI
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
AnkitAI
Model Tree
Qwen/Qwen3-4B
Input Modalities
Output Modalities
Supported Functionality
Recalibrated merge. Same training, better weight blending: +1.8 points on HumanEval-164 over the previous build (74.4 vs 72.6), reproduced across three independent adapters. If you pulled this model before August 2026, re-pull for the stronger build.
<think> on
34% of ordinary prompts and returns nothing. This model answers 34/34 on
the same suite, with 140x less reasoning text and no thinking-mode flag to
manage.Measured on identical harnesses, greedy decoding, Q4_K_M builds, thinking disabled on every row.
| Base Qwen3-4B | This model (v2.1) | |
|---|---|---|
| Prompts answered (34-prompt suite) | 27/34 | 34/34 |
| HumanEval-164 | 79.3 | 74.4 |
| Held-out agent-trace loss | 2.846 | 1.876 |
| BFCL simple_python | 95.3 | 92.3 |
| BFCL multiple | 94.5 | 90.0 |
Take this model for local agent and coding work where you want structured, reliable answers every time: it fits the agent-session distribution far better and never silently returns empty.
Take the base model if your workload is maximum-accuracy function calling in a tool-calling harness, where its few extra points matter more than reasoning style.
Fine-tuned from Qwen/Qwen3-4B (Apache-2.0). Training data: Glint-Research/Fable-5-traces (AGPL-3.0) and Roman1111111/gpt5.5-terminal (MIT). Because those traces originate from third-party assistants, the providers' terms may apply to downstream training and distillation. If you plan to build on this model commercially, confirm your use aligns with those terms.
bibtex
@misc{aglawe2026agenttrace,author = {Aglawe, Ankit},title = {Agent-Trace Fine-Tuning of Small Language Models under Constrained Compute},year = {2026},publisher = {Zenodo},doi = {10.5281/zenodo.21676407},url = {https://doi.org/10.5281/zenodo.21676407}}
The Qwen team for the base model; Glint-Research and Roman1111111 for the trace datasets; empero-ai for the recipe this series iterates on.