GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
redai-infra
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
redai-infra
Model Tree
Qwen/Qwen3-VL-8B-Thinking
Input Modalities
Output Modalities
Supported Functionality
test_all_no_internet, 309 tasks, greedy ×3)| accuracy | input cost | peak context (p95) | |
|---|---|---|---|
| Base operating point (img4, uncompressed) | 33.0% | 100% | 11.5K tok |
| This checkpoint (img2 + skip, compressed) | 37.8% | 53% | −37% |
This is a complete, directly-loadable Qwen3-VL-8B model directory. Deploy it with the compressed observation rule it was trained under — evaluating with the default rich observation underestimates it:
bash
huggingface-cli download redai-infra/hybrid-routing-context-rl --local-dir ckpts/context_rlMODEL=ckpts/context_rl CONTEXT_POLICY=skip_on_mcp_success MAX_IMAGE_HISTORY_LENGTH=2 \ bash scripts/run_mcp_eval.shCode, configs, and full provenance: https://github.com/redai-infra/hybrid-routing-agent — see results/context_rl/PROVENANCE.md for how every number above is derived, and configs/experiments/context_rl.yaml for the exact training recipe.
text
@article{fan2026screenshots, title = {Screenshots or Tools? Eliciting Tool Use and Managing Multimodal Context in Hybrid GUI-MCP Computer-Use Agents}, author = {Fan, Siqi and Li, Minghao and Ma, Xiaoqian and Tan, Wenhui and Huang, Xiusheng and Wu, Juntong and Zhang, Liujie and Shang, Shuo and Chen, Weihang}, journal = {arXiv preprint arXiv:2608.03327}, year = {2026}, url = {https://arxiv.org/abs/2608.03327}}