Laya 1 Benchmarks, Pricing & Rankings
Laya Research · 2026-09-21 · Open · Current
Snapshot 2026-10-14 · updated weekly
Laya 1 ranks #41 of 42 on the AMR leaderboard.
Laya 1 benchmark scores
| GPQA Diamond | 38.4% | |
| MMLU-Pro | 37.6% | |
| SWE-Bench Verified | 15.2% | |
| AIME 2025 | 30.8% | |
| LiveCodeBench v6 | 27.4% | |
| Arena Elo | 1196 |
Laya 1 price, speed & context
Input price
$0.05
per 1M tokens
Output price
$0.20
per 1M tokens
Speed
368
tok/s
Context
64K
tokens
About Laya 1
Laya 1 is a tiny open model for edge agents and browser-local inference, newly tracked this month at the bottom of the quality range.
Models around Laya 1 in the rankings
Laya 1 FAQ
Where does Laya 1 rank?
Laya 1 ranks #41 of 42 tracked AI models on the AMR aggregate as of October 2026, with an aggregate score of 55.4. Its strongest dimension is GPQA Diamond at 38.4%.
How much does Laya 1 cost?
Laya 1 costs $0.05 per million input tokens and $0.20 per million output tokens on the standard API. At a typical 1:3 input/output agent workload, that blends to roughly $0.16 per million tokens.
What is the Laya 1 context window?
Laya 1 supports a 64K token context window and generates at a median 368 output tokens per second. That context fits roughly 43 standard documents in a single prompt.
Is Laya 1 open source?
Yes — Laya 1 ships open weights that you can download and self-host or serve through any compatible inference provider.
What is better than Laya 1?
Celeris-1 ranks one place higher at #40 with an AMR score of 58.2. Laya 1 still wins 0 of the tracked dimensions head-to-head — check the comparison table to see exactly which.