Show HN: Needle: We Distilled Gemini Tool Calling into a 26M Model
221–230 of 255 posts
Re: Show HN: Needle: We Distilled Gemini Tool Calling into a 26M Model
#222A lot of agent workflows really are just tool selection + argument extraction + structured output. How does this behave once workflows become multi-step and state starts accumulating across calls?
Re: Show HN: Needle: We Distilled Gemini Tool Calling into a 26M Model
#223Re: Show HN: Needle: We Distilled Gemini Tool Calling into a 26M Model
#224This seems to be something that could be plugged into an Alexa style device? You had 15 tool categories, what were they and will it work on categories outside of this 15?
Re: Show HN: Needle: We Distilled Gemini Tool Calling into a 26M Model
#225This is really cool to see the push for tiny models
Re: Show HN: Needle: We Distilled Gemini Tool Calling into a 26M Model
#226Nice catch. Using agent for simple tasks is inefficient and wasteful, Needle really resolves this. Looking forward to future upgrades!
Re: Show HN: Needle: We Distilled Gemini Tool Calling into a 26M Model
#227Built a browser version of this runs entirely client-side via onnxruntime-web, no server: https://huggingface.co/spaces/shreyask/needle-playground
Re: Show HN: Needle: We Distilled Gemini Tool Calling into a 26M Model
#228Smaller tool calling models are absolutely needed when we are trying to apply agentic workflow to on-device, small systems, where data is private. I currently use small general purpose LLM. But a special finetune is always the way. I'll try this for my use case.
Re: Show HN: Needle: We Distilled Gemini Tool Calling into a 26M Model
#229Great work Henry!
Re: Show HN: Needle: We Distilled Gemini Tool Calling into a 26M Model
#230What Google's response will be to this.