3 ms·
Also try a hundred bucks on Openrouter and experiment with the Chinese models like Deepseek, Kimi, GLM. $100 will last a good while, and it isn't tied to a sing
by nozzlegear 2mo ago
Also try a hundred bucks on Openrouter and experiment with the Chinese models like Deepseek, Kimi, GLM. $100 will last a good while, and it isn't tied to a single model or provider.
- halfmatthalfcat 2mo agoI burned though $50 on OpenRouter in three days, this is with selective model routing based on tasks/strict AGENTS.md. A Claude 20x sub running Sonnet 5/Opus 5 and light Fable usage gets me through the month. It wasn't even close even though I wish it were.
- nozzlegear 2mo agoDamn that's crazy. I put in $10 on the 5th of August and still have $1.99 left today, the 18th. I use Deepseek Flash almost exclusively.
- nl 2mo agoDeepseek Flash is somewhere around the OpenAI Luna level though. That's not bad (and nor is Luna!) but Opus/Sol are just massively better especially once you start doing agentic tasks. See for example https://x.com/zainhas/status/2085567832268656755 https://x.com/zainhas/status/2085567832268656755
- nozzlegear 2mo agoAgreed, it's not SOTA but I'm happy with the trade off. My preference is to use mostly local models and augment them with Deepseek. It's been a pretty great equilibrium.
- planckscnst 2mo agoI have the $100 plan on OpenCode Zen, which I imagine is similar. I burn through it ridiculously fast. I haven't run the numbers, but it felt about the same as an Anthropic $20 plan
- christophilus 2mo agoYeah. Same. Deepseek Flash goes a long way, but it’s not capable enough for many tasks, so I end up defaulting to my Codex sub and just use Sol.