Back
FR

Model creator

Google

Every Google model in the BenchSift catalogue, with its scores, ranks, prices and capabilities as the sources publish them.

Hugging Face · google
View in the catalogue
Models
125
146 configurations, reasoning levels included
Open weights
27
27 of 125 models
Latest release
Nov 11, 2026
Gemini Decision 1 · First dated model: May 10, 2023
Median price
$1.13
$0.040 to $8 per million tokens, across 28 priced models

Best scores and ranks

The creator's best model on each measure, and its rank among every measured model in the catalogue.

Intelligence

Gemini 4 Carbon (High)
60.0Rank 4 of 781 measured models

Coding

Gemini 3.8 Flash (High)
76.3Rank 18 of 300 measured models

Math

Gemini 3 Flash Preview (Reasoning)
97.0Rank 3 of 270 measured models

Agentic

Gemini 3.8 Flash (High)
40.2Rank 44 of 191 measured models

Output speed

Gemini 3.7 Flash (Medium)
392 t/sRank 5 of 449 measured models

Lowest price

Gemma 4 E4B (Non-reasoning)
$0.040Rank 50 of 740 measured models

OpenRouter ranking

Gemini 3.8 Flash (High)
#20

Capabilities

Number of models that declare each capability. Tool calling and structured outputs come from OpenRouter.
Reasoning54 of 125
Open weights27 of 125
Tool calling18 of 125
Structured outputs29 of 125

Input

  • Text122
  • Image92
  • Video58
  • Audio57
  • Files20

Output

  • Text89
  • Speech21
  • Image16
  • Video15
  • Embeddings5
  • Audio2
  • Transcription2
  • Decisions1

Overview

Model types

  • Text89
  • Audio15
  • Video13
  • Image4
  • Embeddings3
  • Decisions1

Releases by year

  • 202652
  • 202541
  • 202410
  • 20233

Hugging Face

Official repositories confirmed by Hugging Face metadata.

6 official repositories

Licenses

  • gemma5
  • apache-2.01

Models

125

Reasoning levels are folded into one entry. The rank is the whole catalogue's: sorting does not change it.

Where these figures come from

Each catalogue model is tied to its creator by the identity Artificial Analysis publishes, or OpenRouter's and Models.dev's for models it does not track. Counts fold the reasoning levels of one model together. Ranks compare each measure with every measured model in the catalogue, configurations included, as model pages do. The price is the published blended price (3 input tokens to 1 output token).

Figures computed by BenchSift from the catalogue: Artificial Analysis, OpenRouter, Models.dev and Hugging Face. A value no model carries is not shown.

Data retrieved on Nov 12, 2026.

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