I've switched to non-SOTA models, which deliver comparable value at a fraction of the cost. A full day of coding with Deepseek is approx $1 in tokens, and at least for my use cases the quality is equivalent to Claude.

▲

profsummergig a day ago | parent | next [-]

If you don't mind sharing... what's your preferred vendor for Deepseek

	▲	fancy_pantser a day ago \| parent [-]
		Not who you asked, but if you're looking for a rec, I've been using GLM and Deepseek models through Deepinfra because the prices are decent for US-based deployments and there's acceptable throughput and policy/terms. I'd love to find another that lowers prices over time when they roll out e.g. multi-token prediction instead of just taking a larger margin, but I don't know of any today.

▲

christophilus a day ago | parent | prev [-]

Hm. I must be holding it wrong. I hit $20 easily in 5 hours with deepseek in opencode.

	▲	p1necone a day ago \| parent \| next [-]
		Same here - curious which deepseek model and what op was doing. I suspect it might be a question of conversational loop vs agentic dev - the former uses much less tokens than letting an autonomous agent churn away on your codebase.
	▲	fragmede a day ago \| parent \| prev [-]
		Do a through audit of every single token you're sending and receiving. Some harnesses include things you didn't expect.