gaents Dashboard

← Models

gaents/claude-opus

Call it with "model": "gaents/claude-opus" at https://api.gaents.com/v1/chat/completions.

Price and service

Input, per 1M tokens
$4.6420
Output, per 1M tokens
$23.2100
Source
GET https://api.gaents.com/v1/models at 2026-11-19T06:07:16Z. You pay the provider's reported cost plus the margin; the unused part of each quote is refunded.
Serving
yes; live routes 1 (alert set below 2: below it now) (/v1/capacity at 2026-11-19T06:07:32Z)
Calls on gaents
No calls yet in the reliability window (/v1/reliability, last 30 days)
Latency p50 (ok calls)
no successful calls yet

Measured on our tasks

TaskRouteQuality [95% interval]Correct / nCash per correctCapacity per correct (not cash)Source
Coding tasks (agent with a harness) go-toclaude-opus (Claude Code harness) passes 86% [72%-93%]
mergeable 100% [70%-100%]
36 / 42 not in source per passing run $0.9086 swarm experiments/coding-h2h/RESULTS-r2-2026-11-09.md ('Update 17:59Z') + Judge coding-diff-review-2026-11-17, 2026-11-09

Evidence files are not published yet (setting evidence.publish_model_review_full = false); every number is in /models/models.json with its source and date. Re-run: the task inputs are not public, so a public re-run is not offered.

Failover, by family

The router tries each route in this order and skips one that is down, out of balance or failing. Steps are shown by family; providers and exact ids are not shown (show settings).

  1. claude-opus: 3 routes, this alias's own weights
  2. glm: 1 route, a build with no public alias

From the router's routes table (read 2026-11-19T04:27:48Z). Whether a route is live now is the capacity count above.