Eval Library
TA
For Together AIAI Platform

Rate Limiting 429 Recovery

Together AI · Together AI

Together AI evals — Rate Limiting & 429 Recovery (relift v3)

About Together AI

Together AI is an enterprise AI inference cloud providing fast, scalable access to leading open-source models via an OpenAI-compatible API. Teams use Together for production inference, fine-tuning, and dedicated GPU deployments without the complexity of self-managed infrastructure.

Employees

~100

Industry

AI Inference Platform

Headquarters

San Francisco, CA

Sample tests· showing 3 of 9

#InputExpected behaviorCheck
01

Above dynamic rate per docs; must backoff not spin tight loop.

{"criteria": ["Reads x-ratelimit-reset", "Exponential backoff", "Does not confuse with 503"], "pass_threshold": 3}

Pass / FailAi Platformcritical
02

Reduce tokens per request or spread traffic; steady traffic raises limits over time.

["Splits prompt across chunks where policy allows", "Lowers max_tokens", "Spreads requests evenly across minute"]

Pass / FailAi Platformhigh
03

Docs: spread ~1 RPS across minute vs 60 in one second.

["Shapes timer to ~1 req/s", "Monitors success rate improving dynamic limit", "Uses batch for offline bulk"]

Pass / FailAi Platformmedium

Rubric criteria

  • Together Ai
  • Ai Platform
  • Rate Limiting 429 Recovery

Recommended for

Together AITogether AI customers

Works with

Related evals

Run this eval in your workspace

Connect your data, configure thresholds, and review results with your team.