Celeris-1 Benchmarks, Pricing & Rankings
Celeris · 2026-08-11 · Closed · Current
Snapshot 2026-10-14 · updated weekly
Celeris-1 ranks #40 of 42 on the AMR leaderboard.
Celeris-1 benchmark scores
| GPQA Diamond | 42.6% | |
| MMLU-Pro | 41.9% | |
| SWE-Bench Verified | 18.4% | |
| AIME 2025 | 35.2% | |
| LiveCodeBench v6 | 31.8% | |
| Arena Elo | 1208 |
Celeris-1 price, speed & context
Input price
$0.20
per 1M tokens
Output price
$0.80
per 1M tokens
Speed
1431
tok/s
Context
128K
tokens
About Celeris-1
Celeris-1 is the fastest model tracked at 1,431 tok/s output — purpose-built for autocomplete, routing, and sub-100ms agent loops.
Models around Celeris-1 in the rankings
Celeris-1 FAQ
Where does Celeris-1 rank?
Celeris-1 ranks #40 of 42 tracked AI models on the AMR aggregate as of October 2026, with an aggregate score of 58.2. Its strongest dimension is GPQA Diamond at 42.6%.
How much does Celeris-1 cost?
Celeris-1 costs $0.20 per million input tokens and $0.80 per million output tokens on the standard API. At a typical 1:3 input/output agent workload, that blends to roughly $0.65 per million tokens.
What is the Celeris-1 context window?
Celeris-1 supports a 128K token context window and generates at a median 1,431 output tokens per second. That context fits roughly 85 standard documents in a single prompt.
Is Celeris-1 open source?
No — Celeris-1 is a closed model available only through Celeris’s official API and authorized cloud partners.
What is better than Celeris-1?
Mercury 2.5 ranks one place higher at #39 with an AMR score of 62.0. Celeris-1 still wins 0 of the tracked dimensions head-to-head — check the comparison table to see exactly which.