Text Generation
Transformers
Safetensors
qwen3_next
conversational
Eval Results

Good for Roleplay - Sucks at tool calling

#13
by abiteddie - opened

Why is this model so bad for agentic and tool usage? Even gpt oss 20b is better!

maybe because this is the thinking version, while the agentic and tool use is the instruct version?

Sign up or log in to comment