BenchLeader
CelerisAuto-detected

Celeris-1

Not yet ranked: too few independent results so far. Released 24 Jul 2026.

Blended price
$0.325/M
$0.200 in · $0.700 out
Output speed
1360 tok/s
First answer
0.58 s
first token 0.58 s
Context
131k
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Knowledge30
  2. Long context45
  3. Composite37

Benchmark results

One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

BenchmarkdefaultSource
GPQA Diamond (AA)not in index63.1%#340Artificial Analysis
Humanity's Last Exam (AA)not in index6.8%#308Artificial Analysis

Coding

BenchmarkdefaultSource
SciCode (AA)not in index21.6%#157Artificial Analysis

Knowledge

BenchmarkdefaultSource
AA-Omniscience-71.6#441Artificial Analysis

Long context

BenchmarkdefaultSource
AA-LCR38.0%#294Artificial Analysis

Composite

BenchmarkdefaultSource
AA Intelligence Index6.3#449Artificial Analysis

What a task costs

Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.

WorkloadTokens in / outCostWith cachingTime
Chat reply400 / 300$0.00030.8 s
Summarise a 30-page report12,000 / 600$0.00281.0 s
Code edit6,000 / 1,500$0.00231.7 s
Agentic coding session60,000 / 4,000$0.0153.5 s
Structured extraction2,000 / 200$0.00050.7 s

See also

Data as of 9 Sept 2026. Compare with another model.