Ollama and vLLM Add Support for Recursive_SI’s Flint Coding Model

Ollama made Recursive_SI’s Flint coding model available to run locally, while vLLM announced day-zero support with NVFP4 and BF16 ready. Flint is an open mixture-of-experts model with roughly 40 billion parameters; vLLM says 2.5 billion are active per token. Ollama describes it as designed for on-device agentic coding. The platforms reported performance under different hardware and context conditions. Ollama reported more than 125 tokens per second on an M4 Pro MacBook Pro and more than 150 tokens per second on a 64GB NVIDIA RTX Spark, both with short contexts. vLLM reported that its custom NVFP4 kernels delivered 500 tokens per second on a single GB200, even at a 100K-token context, without speculative decoding.

Generated by AI from reports · Updated 47 minutes ago

1. 10/22 01:22 (UTC+8) · 1 reports
   Ollama Adds Recursive Superintelligence’s Flint Model for Local Coding
   Ollama · X：Ollama introduces Flint, Recursive_SI’s ~40B MoE open model for on-device coding
2. 10/22 00:45 (UTC+8) · 1 reports
   vLLM Adds Day-One Support for Flint
   vLLM · X：vLLM adds day-zero support for Recursive_SI’s Flint coding model

There is not enough continuous observation data to plot a trend yet.