Back

Google

Gemini 3.8 Flash

ReasoningClosed weights
Intelligence
33.5
Rank 110 of 749 measured models
Coding
73.5
Rank 38 of 298 measured models
Price/1M
$1.50
Rank 436 of 671 measured models
Context window
1M
Rank 86 of 1216 measured models

Gemini 3.8 Flash (Low) scores 33.5 on the Artificial Analysis Intelligence Index, ranking 110th of 749 measured models. The closest models on this index are GLM-5.2 (33.7), Muse Spark 1.1 (Xhigh) (33.7), and Motif 3 (33.6). At $1.50 per million tokens (3:1 input/output blended price), it is cheaper than GLM-5.2 ($2.15) and Muse Spark 1.1 (Xhigh) ($2).

Artificial Analysis Indices
Low
Intelligence33.5
Rank 110 of 749 measured models
Coding73.5
Rank 38 of 298 measured models
Agentic32.6
Rank 62 of 190 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 ↗Cost per task
Capability indexes
Performance in specific industries, measured by Artificial Analysis
Finance & Accounting39.1
Rank 82 of 224 measured models
Strategy & Ops38.5
Rank 97 of 224 measured models
Legal44.5
Rank 62 of 224 measured models
Engineering34.5
Rank 96 of 223 measured models
Economics45.6
Rank 78 of 231 measured models
Standard Benchmarks
GPQA Diamond92.0%
Rank 43 of 620 measured models
HLE37.1%
Rank 109 of 669 measured models
SciCode55.0%
Rank 54 of 236 measured models
AA-LCR80.7%
Rank 65 of 580 measured models
Terminal-Bench 2.183.1%
Rank 36 of 245 measured models
Terminal-Bench v410.1%
Rank 88 of 223 measured models
τ-Bench Banking33.2%
Rank 58 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