{"slug": "improving-tool-use-with-small-llms", "title": "Improving Tool Use with Small LLMs", "summary": "A developer using the LFM 2.5 2.6B model with llama-cpp found that adding detailed descriptions to JSON schema parameters improved tool call correctness from 0% to nearly 100%, enabling the use of a fast 1.7 GB model without a GPU for an MCP server.", "body_md": "I have been trying to develop an agentic AI system with a **local MCP server**. I am using `llama-cpp`\n\nwith the **LFM 2.5 2.6B** model—this LLM is only 1.7 GB and is very fast. However I found the model used a parameter in one of the tool calls correctly about 0% of the time.\n\nOther models, including Gemma 4, **used the parameter correctly** nearly 100% of the time. So I tried to figure out what the problem was. I determined that parameters in the **JSON schema** need to have a detailed **description**—describing the params in the tool call descriptions was not enough.\n\nHere are my discoveries:\n\n• With descriptions on the parameters specifically, tool calling was correct nearly 100% of the time in LFM 2.5 2.6B.\n\n• It is best not to be shy about explaining how to use a parameter—a 6-line explanation is perfectly fine, and will help smaller models understand better.\n\nIt is kind of **boring** to write a 6-line paragraph about a tool call parameter, but it seems to really **help smaller models**. I went from 0% compliance on a parameter to 100% compliance—and this means I can use a small 1.7 GB model which even runs fast without a GPU for my MCP server.", "url": "https://wpnews.pro/news/improving-tool-use-with-small-llms", "canonical_source": "https://www.dotnetperls.com/2026_8_7_improving-tool-use-small-llms", "published_at": "2026-08-07 07:00:00+00:00", "updated_at": "2026-08-09 12:54:27.465789+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-agents", "ai-tools"], "entities": ["LFM 2.5 2.6B", "llama-cpp", "Gemma 4", "MCP server"], "alternates": {"html": "https://wpnews.pro/news/improving-tool-use-with-small-llms", "markdown": "https://wpnews.pro/news/improving-tool-use-with-small-llms.md", "text": "https://wpnews.pro/news/improving-tool-use-with-small-llms.txt", "jsonld": "https://wpnews.pro/news/improving-tool-use-with-small-llms.jsonld"}}