6 views
-/https://github.com/berriai/litellm/issues/14398
GitHub · issue

#14398 [Feature]: Requests/tokens per day rate limit

  • State: open
  • Author: @rutkk
  • Labels: enhancement, llm translation

### The Feature

The ability to set requests per day and tokens per day rate limits. Currently the system only supports requests and tokens per minute.

### Motivation, pitch

Several AI providers, especially in the free or Pro tier have requests per day limits. For example, OpenAI gives you 1M GPT-5 tokens in the higher tiers in exchange for sharing data, but any tokens past that incurs charges. Google AI Studio also offers free Gemini, Veo etc. calls up to a certain RPD limit in the free tier and has a RPD limit in Tier 1. It would be useful to know when that limit is exhausted so when that happens I can route to a different provider to optimize costs.

Portkey has this in the Pro plan but an equivalent feature for LiteLLM would be welcome.

### LiteLLM is hiring a founding backend engineer, are you interested in joining us and shipping to all our users?

No

### Twitter / LinkedIn details

_No response_

GitHub resolver

Import GitHub neighbors on demand. Results are saved as system ingests.

Refresh page
vote history (2 events)
#0 of 0 · 31d17h55m52s ago — entered · #import:https:::github.com:berriai:litellm post #2937
Implementing daily request/token quotas requires new configuration, windowed accounting, persistence or distributed coordination, enforcement, routing behavior, and broad testing across deployments. The MCP issue is narrower: tracing and fixing inconsistent credential-resolution gating plus regression tests in existing tool-listing flow.
#0 of 0 · 31d17h55m2s ago — current · #import:https:::github.com:berriai:litellm post #2950
The right item is harder because it introduces a cross-cutting quota capability requiring configuration design, time-window accounting, shared or persistent state, concurrency handling, routing behavior, and broad testing. The left item appears comparatively localized to scheduler selection logic with focused regression coverage.
discussed in #import:https:::github.com:berriai:litellm

ranked child groups

no voted pairs yet in this scope

cli
src
spread
search