AI Benchmark Model

Gemini 3.8 Flash (medium)

Benchmark scores, pricing, speed, and model comparisons for Gemini 3.8 Flash (medium).

Key scores

Intelligence56.6Index score
Coding74.1Index score
Blended Cost$1.50Per 1M tokens

Gemini 3.8 Flash (medium) is an artificial intelligence model developed by Google and released on September 2, 2026. This model is evaluated across several performance metrics, achieving an intelligence index of 56.6 and a coding index of 74.1. Benchmark results for the model include a GPQA score of 0.935, an HLE score of 0.421, a SciCode score of 0.544, and an LCR score of 0.82. Regarding cost, the model is priced at $0.75 per million tokens for input and $3.75 per million tokens for output, resulting in a blended rate of $1.50 per million tokens. Specific data regarding math capabilities, output speed, and time to first token remain currently unknown for this release.

Benchmark results

4 tracked benchmark rows for Gemini 3.8 Flash (medium).

GPQA Reasoning
93.5%
LCR Other
82.0%
SciCode Other
54.4%
HLE Other
42.1%

Comments (0)

No comments yet

Be the first to share your thoughts!