AI Benchmark Model
Celeris-1
Benchmark scores, pricing, speed, and model comparisons for Celeris-1.
Key scores
Celeris-1 is an artificial intelligence model developed by Celeris and released on July 24, 2026. The model demonstrates an intelligence index of 11.8 and a coding index of 14.4, while its math index remains unknown. In standardized testing, Celeris-1 achieved a GPQA score of 0.631, an HLE score of 0.066, a SciCode score of 0.207, and an LCR score of 0.27. Regarding operational performance, the model provides an output speed of 2116.311 tokens per second with a time to first token of 0.46 seconds. Pricing for the model is set at $2.00 per million tokens for input and $6.00 per million tokens for output, resulting in a blended cost of $3.00 per million tokens.
Benchmark results
5 tracked benchmark rows for Celeris-1.
MMLU Pro Knowledge
78.0% GPQA Reasoning
63.1% LCR Other
38.0% SciCode Other
21.6% HLE Other
6.8%