Comment on Ask HN: A good Model to choose in Ollama to run on Claude CodeComments−parthsareen7moHey! One of the maintainers of Ollama. 8GB of VRAM is a bit tight for coding agents since their prompts are quite large. You could try playing with qwen3 and at least 16k context length to see how it works.
Comments
Hey! One of the maintainers of Ollama. 8GB of VRAM is a bit tight for coding agents since their prompts are quite large. You could try playing with qwen3 and at least 16k context length to see how it works.