Back

Amazon

Nova 2.0 Pro Preview

ReasoningClosed weights
Intelligence
12.8
Rank 351 of 744 measured models
Coding
25.9
Rank 198 of 298 measured models
Output speed
120 t/s
Rank 131 of 403 measured models
Price/1M
$3.78
Rank 529 of 659 measured models
Context window
256K
Rank 518 of 1199 measured models

Nova 2.0 Pro Preview (Low) scores 12.8 on the Artificial Analysis Intelligence Index, ranking 351st of 744 measured models. The closest models on this index are Nemotron 3.5 Lightning (12.9), GLM-4.5 (Reasoning) (12.8), and Nemotron 3 Super 120B A12B (12.8). At $3.78 per million tokens (3:1 input/output blended price), it is more expensive than Nemotron 3.5 Lightning ($0.095) and Nemotron 3 Super 120B A12B ($0.45). With an output speed of 120 t/s, it is slower than Nemotron 3.5 Lightning (302 t/s) and Nemotron 3 Super 120B A12B (146 t/s).

Artificial Analysis Indices
Low
Intelligence12.8
Rank 351 of 744 measured models
Coding25.9
Rank 198 of 298 measured models
Math63.3
Rank 111 of 270 measured models
Reasoning levels
Each point is a reasoning level of this model: more reasoning often raises the score but costs more. Select a point to open that level.
Intelligence Indexmost efficient ↗Price / 1M tokens
Standard Benchmarks
MMLU Pro82.2%
Rank 73 of 349 measured models
GPQA Diamond75.1%
Rank 269 of 619 measured models
HLE5.2%
Rank 463 of 664 measured models
LiveCodeBench63.8%
Rank 104 of 343 measured models
AIME 202563.3%
Rank 111 of 270 measured models
IFBench79.6%
Rank 11 of 450 measured models
AA-LCR64.7%
Rank 258 of 575 measured models
TerminalBench Hard17.4%
Rank 189 of 432 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