GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
trillionlabs
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
trillionlabs
Model Tree
Qwen/Qwen3-4B-Thinking-2507
Input Modalities
Output Modalities
Supported Functionality
| Property | Value |
|---|---|
| Base model | Qwen3-4B-Thinking-2507 |
| Checkpoint | Global step 120 |
| Training | verl GRPO with SEAL |
| Training questions | ~1,613 value-answer items in difficulty band [1, 7] |
| Environment | Nine tools over ~1.12M computer-science papers |
The checkpoint was evaluated with the native AstaBench tool surface on the
full-text-available LitQA2 test subset (n=75).
| Model | Accuracy |
|---|---|
| Qwen3-4B-Thinking-2507 base | 0.173 |
| S3 Qwen3 4B, step 120 | 0.453 |
Accuracy increased through step 120 and declined at later checkpoints. Transfer to the AstaBench paper-finder task remained approximately flat, suggesting that the gain is specific to QA-like tasks.
Step 120 was selected after inspecting this test-set checkpoint curve. The reported result is diagnostic rather than an unbiased final benchmark.
python
from transformers import AutoModelForCausalLM, AutoTokenizermodel_id = "trillionlabs/sim-scholar-qwen3-4b"tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained(model_id,device_map="auto",torch_dtype="auto",)
Reproducing the reported behavior requires compatible tool definitions, prompting, observation formatting, and stopping behavior. These components are available in the linked code repository.
This checkpoint is intended for research on literature-search agents, tool-using language models, agentic reinforcement learning, and sim-to-real transfer.
The checkpoint should not be used as an authoritative source of scientific or medical information.
This checkpoint is licensed under the Apache License 2.0, consistent with the base model. Training-data and third-party material retain any applicable original terms.