V4 flash is much worse than any Claude model. If you're doing something simple, it can be a good way to save money though.
throwa356262 1 hours ago [-]
I agree that Claude is better (definitely better than the flash version which is relatively small). But...
I actually canceled my Claude Code plan a few months back after trying out some of the "lesser" models on openrouter. They seem to work as just as well (or just as bad) for my coding tasks.
But China?
https://www.reuters.com/world/china/openai-accuses-deepseek-...
From HF: 284B parameters (13B active), 1M context window.
This is indeed some kind of compressed context and the quality goes down as the context grows. IIRC the V4 paper had some numbers on this
https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash
I actually canceled my Claude Code plan a few months back after trying out some of the "lesser" models on openrouter. They seem to work as just as well (or just as bad) for my coding tasks.