{"slug": "flint-llm-in-pi-on-rust-code", "title": "Flint LLM in Pi on Rust Code", "summary": "A hands-on test found that Recursive Superintelligence's Flint model, released last month with 2.5 billion active parameters out of 40 billion total, correctly removed all unsafe code from a Rust program and added the forbid_unsafe directive, using 30,000 tokens at 40 tokens per second on a local llama-cpp router. The tester reported the modified Rust program compiled correctly and that Flint used project-specific utility functions, though the unsafe-free version ran much slower and was discarded. The tester concluded Flint (Flint-Q4_K_M.gguf) is likely a stronger coding model than Laguna XS 2.1 and Qwen 3.6 35B, making it their first choice for local LLM coding.", "body_md": "I have been meaning to test out the **Flint model** from Recursive Superintelligence on my **Rust program** in Pi code. Flint was released last month, and has impressive benchmarks (it outperforms Laguna XS 2.1 and Qwen 3.6 35B). It is also fast as it only uses 2.5 billion active parameters (of 40 billion total).\n\nAfter some issues (aren't there always issues) I got Pi working with `llama-cpp` on the **local model router**. I had Flint remove all `unsafe` code in my Rust program and add the `forbid_unsafe` directive. It took 30,000 tokens and proceeded at a nice pace (40 tokens per second), and my PC handled the load easily.\n\nIt correctly made the required changes and even used the **utility functions** specific to my project. So I was impressed, both on Flint's speed and the quality of its output. The Rust program compiled correctly. Unfortunately, the version with no `unsafe` code is much slower, so I discarded it—but this was not the fault of the LLM, it was just an experiment.\n\nI am inclined to believe in Recursive's benchmarks: **Flint** (`Flint-Q4_K_M.gguf`) is probably a **stronger coding model** than the Laguna or Qwen models it was compared against. And this means it is likely my first choice for local LLM coding now.", "url": "https://wpnews.pro/news/flint-llm-in-pi-on-rust-code", "canonical_source": "https://www.dotnetperls.com/2026_11_12_flint-llm-pi-rust-code", "published_at": "2026-11-12 07:00:00+00:00", "updated_at": "2026-11-13 15:51:03.477449+00:00", "lang": "en", "topics": ["large-language-models", "ai-tools", "developer-tools"], "entities": ["Recursive Superintelligence", "Flint", "Flint-Q4_K_M.gguf", "Laguna XS 2.1", "Qwen 3.6 35B", "Rust", "llama-cpp"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/flint-llm-in-pi-on-rust-code", "markdown": "https://wpnews.pro/news/flint-llm-in-pi-on-rust-code.md", "text": "https://wpnews.pro/news/flint-llm-in-pi-on-rust-code.txt", "jsonld": "https://wpnews.pro/news/flint-llm-in-pi-on-rust-code.jsonld"}}