About this vendor
Vendor-level selection advice by Y-API. The per-model pages linked below carry the model-specific checks; nothing here is a measured comparison.
Where this vendor fits
The Claude entries are what the catalog’s cheaper models get compared against: an order of magnitude above the Flash tier on price, so the practical question is whether a workload measurably improves at that cost, not whether the models are capable.
What to watch for
Neither entry has a public model card to audit, and migration from older Claude models involves a new tokenizer and a different accepted parameter set — the reviewed Sonnet entry does not list temperature. File input upstream does not prove Y-API forwards documents correctly.
Models on Y-API
Prices are account-credit snapshots from the Y-API catalog. Each model links to its own guide with a cost estimator and a request example.
| Model | Context window | Input / 1M tokens | Output / 1M tokens |
|---|---|---|---|
Claude Opus 5anthropic/claude-opus-5 | 1,000,000 | $5.00$0.50 | $25.00$2.50 |
Claude Sonnet 5anthropic/claude-sonnet-5 | 1,000,000 | $2.00$0.20 | $10.00$1.00 |
Account credit / Cash equivalent · $1 paid adds $10 of credit; the cash equivalent is credit consumed ÷ 10. These are Y-API prices, not the publisher’s list prices.
Choosing within the lineup
Default to Sonnet 5 for day-to-day pull requests and document analysis. Hold Opus 5 for second-pass reviews of consequential changes, and only after a comparison shows it catches defects Sonnet misses — escalation should be measured, not habitual.
Before you choose
Do Claude entries on Y-API differ from Anthropic’s own API?
The models are Anthropic’s; the gateway is not an Anthropic endpoint. Parameter forwarding and document handling have their own compatibility boundaries, and requests must use the exact Y-API catalog IDs rather than Anthropic’s native model names.
Sources & scope
Vendor scope on this site is exactly the reviewed model guides listed above; a vendor page exists only while at least one of its models is reviewed. Claims about architecture or positioning follow the cited pages. We do not publish measured latency, uptime or benchmark scores for any vendor.