GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
CharlieLLL
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
CharlieLLL
Model Tree
Qwen/Qwen3-1.7B
Input Modalities
Output Modalities
Supported Functionality
These are worker-under-orchestrator results, not solo-model accuracy. Concurrency is 64 top-level episodes. BrowseComp: 150 questions, 3 nodes, TP4 orchestrator and four independent TP1 workers. DR9K: 256 questions, 2 nodes, TP4 orchestrator and TP1/native DP4 workers. Full-answer repository manual grading protocol; not the canonical Gemini semantic judge. Empty/error episodes count wrong.
| Orchestrator | BrowseComp | DR9K |
|---|---|---|
| MiniMax-M2.7 | 98/150 (65.33%) | 110/256 (42.97%) |
| DeepSeek-V4-Flash | 96/150 (64.00%) | 147/256 (57.42%) |
| Nemotron Ultra | 99/150 (66.00%) | 143/256 (55.86%) |
| MiMo-V2.5 | 104/150 (69.33%) | 142/256 (55.47%) |
| Nemotron Super | 88/150 (58.67%) | 119/256 (46.48%) |
| Inkling-Small | 99/150 (66.00%) | 152/256 (59.38%) |
Complete report includes latency settings, difficulty scores, observed role tokens/cache and separate theoretical worker prefix reuse.
BrowseComp traces · DR9K traces.
Use the report's evaluator, benchmark-specific chat templates, search tools and budgets to reproduce the worker results; generic chat generation alone does not reproduce this setup.