In my experience, it is claude-code paired with deepseek-v4. For penny-pinchers like me, I can have long coding sessions with it with no anxiety about the cost. Also, prompting it to what you want and verifying the outputs is more important than the quality of the model. So, I am better off with a cheaper model and taking the responsibility for prompting it and verifying the results.
replyIt's obviously much cheaper paying by the token but how does it compare to a codex subscription on cost?
replyCan you quantify the actual costs in a week and the use you make?
replyNot GP, but for my use I'd estimate $0.10-0.30 per hour of use per agent with DeepSeek v4 Pro
replyCursor with Composer 2.5 seems to be competitive with frontier models (Opus and GPT-5.5) for a significant price discount. Benchmarks are gamed, as always, but $0.55/task vs $11.02 a task definitely indicates that there's some cost advantage.
replyI've gone through ~600m tokens in Xiaomi Mimo though Claude, and it's been the most effective use of an agent I've had yet. It's very capable, but generally not ambitious, picking simple but effective solutions to most problems I give it.
Going to write something longer about the experience when I get to a billion tokens.
replyI do have my eyes on the coding plan, which is quite generous.
replyAre you using Mimo 2.5 pro?
replyYes. I tried a couple of weeks with non-Pro, and it was pretty good, but I had too many spare tokens, so I switched back to Pro. :)
replySeems to be DeepSeek.
replyAlthough I have little interest in agentic coding, when I do use it, I have found Kimi K2.6 to give Opus-quality output, and have switched entirely to it for pretty much everything.
replyI've used Opus extensively and tried K2.6 on a few projects, and the gap is huge. K2.6 is nowhere near the performance of Opus. That's fine because it's also far cheaper, but public benchmarks line up with my own personal experience that they aren't comparable in terms of intelligence.
reply(that is, different places on the Pareto efficiency graph)
No two uses are alike, I suppose. For me, whatever difference is a wash. However, I probably tend to shy away from throwing high-complexity/long-horizon tasks at the model.
replyKimi 2.6 is great. Qwen3.7-max benchmarks similarly but I havent used it yet
replyFor me, it's by far Deepseek. It's many times cheaper than competitors, and about as good as Sonnet 4.6.
replyI'd generally agree about Deepseek being as good as Sonnet - but I have extreme trouble with prompt compliance with V4 Pro in a way that I've never had with Sonnet. I'll tell it "find the bug, but don't fix it" or "please use this tool I just developed" and it'll ignore me a high fraction of the time.
replyIt's bad enough that I'm working on guardrails at the harness level because prompting appears to be useless.
Do you have the same issue?
I have Opus make a fairly detailed plan, then Deepseek implements, and GPT reviews. With that setup, I have zero issues, probably because what you mention is handled (the plan keeps it on track and the reviewer catches any issues).
replyNow that you mention it, though, I have seen it do a few things that weren't in the plan. The reviewer caught them, though, so they didn't cause a problem, and it's so cheap that overall it's a massive improvement.
Just use codex with 5.5 on low reasoning levels
reply