Unless I specifically use nemotron-3-super:cloud (and probably other cloud models), offline models just do not work. First off I tried Gemma 4, it said 4 things to me
- I
- Starting
- To
- Under
(and other times i tried it gave me more random words)
then it just throws a
API Error: Claude's response exceeded the 32000 output token maximum. To configure this behavior, set the
CLAUDE_CODE_MAX_OUTPUT_TOKENS environment variable.
``` at me. Right now I am testing qwen3.5:9b
Unless I specifically use nemotron-3-super:cloud (and probably other cloud models), offline models just do not work. First off I tried Gemma 4, it said 4 things to me
(and other times i tried it gave me more random words)
then it just throws a