返回
原文
Chubby♨️@kimmonismus
41AI 编辑部评分，满分 100
2026-08-04 16:22· 19分钟前
跳到正文
AI 摘要

Celeris-1 被 Artificial Analysis 评为输出速度第一，约 2,086 tokens/秒，MMLU-Pro 得分 75.9%，且无需定制推理芯片。

Crazy： Artificial Analysis currently ranks Celeris-1 #1 for output speed at roughly 2，086 tokens per second @ 75.9% on MMLU-Pro！

Celeris says it achieves this on off-the-shelf GPUs， without custom inference silicon.

In Celeris' own same-harness evaluation， the model scored 75.9% on MMLU-Pro， compared with 78.5% for GPT-5 mini and 81.9% for GPT-5， while responding 13-16× faster.

CelerisCeleris-1 is now ranked as the #1 fastest AI model according to @ArtificialAnlys. 2,038 output tokens per second, #1 of all 591 models in response time, TTFT an...
推理评测/基准
在 X 查看原推导出 Markdown
返回
41导出 Markdown
Chubby♨️ · @kimmonismus · X·2026-08-04 16:22·19分钟前
在 X 看原推· x.com
AI 摘要

Celeris-1 被 Artificial Analysis 评为输出速度第一，约 2,086 tokens/秒，MMLU-Pro 得分 75.9%，且无需定制推理芯片。

Crazy： Artificial Analysis currently ranks Celeris-1 #1 for output speed at roughly 2，086 tokens per second @ 75.9% on MMLU-Pro！

Celeris says it achieves this on off-the-shelf GPUs， without custom inference silicon.

In Celeris' own same-harness evaluation， the model scored 75.9% on MMLU-Pro， compared with 78.5% for GPT-5 mini and 81.9% for GPT-5， while responding 13-16× faster.

CelerisCeleris-1 is now ranked as the #1 fastest AI model according to @ArtificialAnlys. 2,038 output tokens per second, #1 of all 591 models in response time, TTFT an...
推理评测/基准
在 X 查看原推x.com