Back

Sapiens AI

Agnes 2.5 Pro Beta

ReasoningClosed weights
Intelligence
35.2
Rank 90 of 747 measured models
Coding
62.3
Rank 77 of 298 measured models
Output speed
118 t/s
Rank 139 of 403 measured models
Price/1M
$0.15
Rank 128 of 660 measured models
Context window
1M
Rank 84 of 1203 measured models

Agnes 2.5 Pro Beta scores 35.2 on the Artificial Analysis Intelligence Index, ranking 90th of 747 measured models. The closest models on this index are DeepSeek V4 Pro 0813 (36.0), Agnes 3.0 Flash (35.5), and DeepSeek V4 Flash Vision (Max) (34.8). At $0.15 per million tokens (3:1 input/output blended price), it is cheaper than DeepSeek V4 Pro 0813 ($1.98) and DeepSeek V4 Flash Vision (Max) ($0.66), more expensive than Agnes 3.0 Flash ($0.075). With an output speed of 118 t/s, it is faster than DeepSeek V4 Pro 0813 (96.3 t/s), slower than DeepSeek V4 Flash Vision (Max) (173 t/s).

Artificial Analysis Indices
Intelligence35.2
Rank 90 of 747 measured models
Coding62.3
Rank 77 of 298 measured models
Standard Benchmarks
GPQA Diamond90.5%
Rank 61 of 620 measured models
HLE37.5%
Rank 103 of 667 measured models
SciCode48.8%
Rank 121 of 234 measured models
AA-LCR83.0%
Rank 25 of 578 measured models
Terminal-Bench 2.169.7%
Rank 76 of 245 measured models
τ-Bench Banking35.7%
Rank 47 of 211 measured models
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy