AI Agent StorePerplexity · No longer listed on Gateway
Perplexity Decider v1.1 27B · Gateway
This older Gateway offer is no longer in the current catalog. For a new project, start with Decider v1.2 and test it on decisions you have already labeled.
First reviewed 2026-10-15 · Checked 2026-11-03. Current access, prices and serving limits are unconfirmed. The details below describe the last listed offer.
We still need comparable independent tests. You can compare the confirmed features and limits below, but we cannot yet give this version an Agent Fit or Practical Value rating. Test it on records with known answers, including cases that should go to a person.
Previously documented features
Token prices measure the small pieces of text the model reads or produces. Capacity is the amount the service accepts in a request; check the limits below for any fallback to another model. A dash means the rate is unverified.
Exact AI Gateway route ID: perplexity/pplx-decider-v1.1-27b. The former Gateway offer accepted text and images for classification, routing and scoring. Downloaded v1.1 weights remain a separate way to use the older model.
Gateway removed this exact ID from its October 18 catalog. Perplexity’s native API redirects older v1 and v1.1 names to v1.2. Its response echoes the requested name, so the name alone does not prove which version ran.
Gateway last listed 262,144 context tokens on October 17. That is a historical hosting limit; the downloadable implementation uses 8,192 tokens.
Output: Yes/no probabilities, choices and ordered scores. Compare decision models, limits and input costs ↗
The last observed Gateway price was $0.02 per million input tokens with free output on October 17. Current availability and pricing are unconfirmed, so this offer is excluded from the cost calculator. Include retries when comparing the total cost.
Provider release and specifications ↗
Check availability and plans ↗
Test a current alternative
Give each model the same examples, tools and definition of success. Record correct outcomes, response time, retries and total cost. Check whether confidence scores match the answers you can verify.
Compare caching and long-context costs ↗