Rate Limit
under review
K
Ken Ng
+1, would love to know the rate limits on deepseek-v3-0324.
Also, I would be happy to deposit credits if it meant I could get higher rate limits (I think several other big severless providers define "spending tiers" with assocated rate limts).
M.R.
Ken Ng
Hi we are actively looking into our Rate Limiting Strategies and how we want to implement them.
K
Ken Ng
M.R. Just wondering if you have an update? I love the latency and throughput offered by Parasail, but we cannot switch our production traffic over without transparency on rate limits and a clear path towards quota increases...
M.R.
Ken Ng: We are working om rate limits, we will have more information to discuss shortly on the topic as its part of this cycle of planning and releases.
This post was marked as
under review
M.R.
planned
We are planning to show the rate limits on all the models.
M.R.
Hi 405b right now is at 20RPM and 70b is at 110 RPM. Can you give more context on how many RPM you are doing?
A
Anonymous
M.R. Don't really remember, but I butted up against 405 once, assumed it was 6/12/24 hours similarly to claude/chatgpt, and went for the 70b. It's good to know it's only a minute.
M.R.
under review