Roboflow

Claude Sonnet 5.5 vs Gemini 3.8 Flash

Compare Claude Sonnet 5.5 and Gemini 3.8 Flash side-by-side.

Compare Claude Sonnet 5.5 vs Gemini 3.8 Flash live

Run the same image across every model that supports a task and compare their outputs side-by-side.

These models don't share enough common tasks for a side-by-side demo. See the comparison table below for their capabilities.

Models in this comparison

Claude Sonnet 5.5 vs Gemini 3.8 Flash on Vision Evals

Claude Sonnet 5.5 scores higher on 3 of the six Vision Evals tasks.

The widest gap is Data Extraction, where Gemini 3.8 Flash leads 97.3% to 90.7%.

Overall, Claude Sonnet 5.5 averages 83.8% (#7 of 60) against 85.1% (#5 of 60) for Gemini 3.8 Flash.

Gemini 3.8 Flash is cheaper ($0.0033 vs $0.0065 per sample), while Claude Sonnet 5.5 is faster (10.8s vs 11.6s per sample).

Claude Sonnet 5.5Gemini 3.8 Flash

Claude Sonnet 5.5 vs Gemini 3.8 Flash Comparison Table

Evals updated September 28, 2026Pricing updated September 28, 2026

PropertyClaude Sonnet 5.5Gemini 3.8 Flash
OrganizationAnthropicGoogle
Categoryclosedclosed
Modalitymultimodalmultimodal
Release DateSep 2026Sep 2026
Context Window1.0M1.0M
Parameters
LicenseProprietaryProprietary
Pricing per 1M tokens
Input $/1M$0.750
Output $/1M$3.75
Vision Tasks
CaptioningDemo
Chart Question Answering
ClassificationDemo
Document Question Answering
Image Tagging
Multi-Label Classification
Object DetectionDemo
OCRDemo
Vision Language
Visual Question AnsweringDemo
Model Features
Foundation Vision
LLMs with Vision Capabilities
Multimodal Vision
Vision Evalsground-truth scores across 6 vision tasks, pooled at low effort
Overall
83.8%
85.1%
Avg cost / sample$0.0065$0.0033
Avg speed / sample10.78s11.65s
By task
Object Detection (low)
74.3%
±0.9, Mean of 3 runs, range 73.5 to 75.3
$0.0098
68.1%
±0.8, Mean of 3 runs, range 67.3 to 69.0
$0.0041
Object Detection (high)
76.8%
±0.4, Mean of 3 runs, range 76.5 to 77.3
$0.014
74.8%
±1.1, Mean of 3 runs, range 73.4 to 75.6
$0.021
Counting (low)
79.3%
±0.7, Mean of 3 runs, range 78.4 to 79.7
$0.0042
78.8%
±2.7, Mean of 3 runs, range 75.7 to 81.1
$0.0036
Counting (high)
82.9%
±1.4, Mean of 3 runs, range 81.1 to 83.8
$0.0053
79.3%
±1.4, Mean of 3 runs, range 78.4 to 81.1
$0.024
Identification (low)
91.7%
±3.1, Mean of 3 runs, range 87.5 to 93.8
$0.0029
97.9%
±1.6, Mean of 3 runs, range 96.9 to 100.0
$0.0014
Identification (high)
90.6%
±0.0, Mean of 3 runs, range 90.6 to 90.6
$0.0033
96.9%
±0.0, Mean of 3 runs, range 96.9 to 96.9
$0.0034
OCR (low)
90.6%
±0.9, Mean of 3 runs, range 90.0 to 91.7
$0.0079
87.3%
±0.8, Mean of 3 runs, range 86.5 to 88.2
$0.0024
OCR (high)
90.9%
±1.5, Mean of 3 runs, range 89.2 to 92.3
$0.011
88.8%
±0.7, Mean of 3 runs, range 88.0 to 89.4
$0.064
Data Extraction (low)
90.7%
±1.5, Mean of 3 runs, range 89.7 to 92.8
$0.0033
97.3%
±0.5, Mean of 3 runs, range 96.9 to 97.9
$0.0018
Data Extraction (high)
93.1%
±0.5, Mean of 3 runs, range 92.8 to 93.8
$0.0036
94.8%
±1.0, Mean of 3 runs, range 93.8 to 95.9
$0.0089
Reasoning (low)
76.4%
±0.7, Mean of 3 runs, range 75.5 to 76.8
$0.0049
81.2%
±0.3, Mean of 3 runs, range 80.8 to 81.5
$0.0034
Reasoning (high)
83.9%
±1.7, Mean of 3 runs, range 82.1 to 85.4
$0.0061
84.5%
±1.0, Mean of 3 runs, range 83.4 to 85.4
$0.021

Claude Sonnet 5.5 vs Gemini 3.8 Flash: Overview

Claude Sonnet 5.5

Claude Sonnet 5.5 is a proprietary multimodal language model from Anthropic and the second release in the Claude 5.5 family, following Claude Opus 5.5. It accepts interleaved text and image input and returns text, operating with a 1M token context window and a maximum output of 128K tokens per request. The model uses adaptive thinking by default, allocating variable reasoning effort per request rather than exposing a manual extended thinking toggle, and its training data cutoff is June 2026. Anthropic positions it as a faster, lower cost complement to Opus 5.5 for well scoped everyday tasks, bug fixing, and producing documents, slides, and spreadsheets.

On visual and agentic evaluations reported at launch, Sonnet 5.5 scores 61.6% on Chartography, a chart recognition test, compared with 15.6% for Claude Sonnet 5, and 80.1% on OSWorld 2.1, a computer use benchmark measuring screenshot driven control of a desktop environment, compared with 57.0% for Sonnet 5. It reports 70.6% on Terminal-Bench 4.0 for agentic coding. Anthropic describes it as the first Sonnet model able to complete Pokemon Red from screenshots alone, and it generates output more than 30% faster than Sonnet 5 while using fewer tokens for equivalent work.

Gemini 3.8 Flash

Gemini 3.8 Flash is a natively multimodal reasoning model in Google's Gemini 3 series, positioned as the speed and cost oriented Flash tier while targeting long-horizon software engineering, autonomous agents, and enterprise workflows. It accepts text, images, video, audio, and PDF documents in a single request and returns text, with an input limit of 1,048,576 tokens and an output limit of 65,536 tokens. Thinking is configurable at low, medium, and high levels, and the model supports function calling, code execution, structured outputs, context caching, search and Maps grounding, file search, and computer use in preview. Image generation, audio generation, and the Live API are not supported.

On vision oriented evaluations the model reports 86.2% on CharXiv Reasoning for chart and figure synthesis and 87.8% on LVBench for long video understanding in agentic mode, alongside 90.8% on Terminal-Bench 2.1 and 61.6% on SWE-Bench Pro for coding. Following Gemini API conventions, it can localize objects by emitting bounding boxes as [ymin, xmin, ymax, xmax] integers normalized to a 0 to 1000 range, which supports prompt driven detection and grounding in addition to captioning, document parsing, and visual question answering. The knowledge cutoff is March 2026, though coverage in some domains reflects the January 2025 cutoff shared across the Gemini 3 family.

Frequently Asked Questions

On Roboflow's Vision Evals, Gemini 3.8 Flash performed slightly better overall. The two split the six vision tasks 3 to 3, but Gemini 3.8 Flash averages 85.1% (#5 of 60) against 83.8% (#7 of 60) for Claude Sonnet 5.5. These scores measure vision capabilities only; either model may perform differently on work outside this benchmark's tasks.

No. On the Vision Evals Data Extraction benchmark at low effort, Gemini 3.8 Flash leads with 97.3% against 90.7%. This is the widest gap between the two models across the benchmark's tasks.

Gemini 3.8 Flash is cheaper to run. Across Roboflow's Vision Evals task mix it averaged $0.0033 per sample against $0.0065. Actual costs depend on your image sizes, prompts, and output length.

Claude Sonnet 5.5 is faster. Across Roboflow's Vision Evals it averaged 10.8s per inference against 11.6s. Latency varies with image size, prompt length, and provider load, so treat these as relative rather than guaranteed figures.