head-to-head

Sakana: Sakana Namazu vs Google: Gemini 3.5 Flash Lite

Side-by-side comparison of specs, pricing, benchmark scores, and task rankings. Updated 2026-09-27.

Sakana: Sakana Namazu Google: Gemini 3.5 Flash Lite
Vendorsakanagoogle
Quality Score100100
Benchmark Score-34.2
Input Price$0.95/M$0.30/M
Output Price$4.00/M$2.50/M
Context Window262,1441,048,576
Max Output65,53665,536
Tool Calling✓✓
Structured Output✓✓
Reasoning Mode✓✓
Vision✓✓
Audio-✓
Benchmark Scores
ai_index-36.6

The verdict

For a sample job of 1 million input tokens and 250,000 output tokens, Gemini 3.5 Flash Lite costs $0.93 and Sakana Namazu costs $1.95, so Gemini 3.5 Flash Lite is about 2.1 times cheaper for the same work.

Gemini 3.5 Flash Lite reads up to 1,048,576 tokens at once, roughly 1,573 pages of text, against about 393 pages for Sakana Namazu.

Only Gemini 3.5 Flash Lite offers audio input and video input.

Across the 20 tasks where they differ, Gemini 3.5 Flash Lite on 20 (SQL Generation, Code Review, Code Completion and others).

Sakana Namazu was released in August 2026 and Gemini 3.5 Flash Lite in July 2026.

Pick Gemini 3.5 Flash Lite if you need lower cost, long documents, audio input, video input or the stronger all-round task scores.

Costs use current list prices per million tokens; page counts assume about 0.75 words per token and 500 words per page.

Who wins by task?

TaskSakana: Sakana NamazuGoogle: Gemini 3.5 Flash Lite
SQL Generation 130 138
Code Review 126 139
Code Completion 128 132
Code Refactoring 127 141
Bug Fixing 130 143
Unit Test Generation 121 128
Code Documentation 125 134
Regex Writing 119 123
CI/CD Pipelines 117 124
Frontend Component Design 122 126
Data Analysis 124 129
CSV / Spreadsheet Cleanup 128 135
ETL Scripting 122 133
JSON Extraction 131 131
Bulk Data Labeling 128 129
OCR / Document Parsing 128 132
Table Extraction from PDFs 128 132
Long-Document Summarization 129 143
Short-Form Summarization 123 125
Blog Post Writing 118 126

Scores reflect capability match + benchmark data + pricing for each task. Methodology →

Related comparisons

Anthropic: Claude Opus 5 (batch) vs Google: Gemini 3.5 Flash Lite ByteDance Seed: Seed-2.0-Code vs Google: Gemini 3.5 Flash Lite ByteDance Seed: Seed-2.0-Code vs Sakana: Sakana Namazu ByteDance Seed: Seed 2.1 Turbo vs Google: Gemini 3.5 Flash Lite ByteDance Seed: Seed 2.1 Turbo vs Sakana: Sakana Namazu DeepSeek: DeepSeek Flash Latest vs Google: Gemini 3.5 Flash Lite DeepSeek: DeepSeek Flash Latest vs Sakana: Sakana Namazu DeepSeek: DeepSeek V4.1 Flash vs Google: Gemini 3.5 Flash Lite

The Model Movers Report

One email every Friday, built from this site's own rankings: the current top five by benchmark score, every model released in the last seven days, and one note worked out from that week's numbers. You can unsubscribe from any issue with one click.