LLM Comparison
Gemini 3.6 Flash vs Sonar Deep Research
Side-by-side specs, pricing & capabilities · Updated September 2026
Price vs Intelligence
Add to comparison
2/6 models| Organization | ||
| OpenTools Score | 54 12.0 | 25 5.1 |
| Family | Gemini | Sonar |
| Status | Current | Current |
| Release Date | Jul 2026 | Mar 2025 |
| Context Window | 1.0M tokens | 128K tokens |
| Input Price | $1.50/M tokens | $2.00/M tokens |
| Output Price | $7.50/M tokens | $8.00/M tokens |
| Pricing Notes | DeepMind model page lists $1.50 input and $7.50 output per 1M tokens with no caching in the performance table. The ai.google.dev API docs were attempted through ScrapingBee and returned HTTP 400, so exact billing details beyond the official DeepMind price table were not used. | — |
| Capabilities | textvisionaudio-inputvideo-inputpdf-inputreasoningcodingtool-usesearch-groundingcomputer-usestructured-output | textcode |
| Max Output | 64K tokens | — |
| API Identifier | gemini-3.6-flash | perplexity/sonar-deep-research |
| Benchmarks | ||
| Artificial Analysis Intelligence Index | 50artificial-analysis | — |
| SWE-Bench Pro | 58.7google-deepmind | — |
| Terminal-Bench 2.1 | 78google-deepmind | — |
| GDPVal-AA v2 | 1421google-deepmind | — |
| Artificial Analysis Intelligence Index v4.3 | 34https://artificialanalysis.ai/models/gemini-3-6-flash | — |
| GPQA Diamond | 92.8283artificial-analysis | 47.1perplexity |
| SciCode | 53.3565artificial-analysis | — |
| MMLU Pro | — | 68.9perplexity |
| MATH-500 | — | 81.7perplexity |
| AIME | — | 48.7perplexity |
| View Gemini 3.6 Flash | View Sonar Deep Research | |
Cost Calculator
Enter your expected monthly token usage to compare costs.
| Model | Input | Output | Total / mo | vs Best |
|---|---|---|---|---|
| Gemini 3.6 FlashCheapest | $1.50 | $3.75 | $5.25 | — |
| Sonar Deep Research | $2.00 | $4.00 | $6.00 | +14% |
Gemini 3.6 Flash
Gemini 3.6 Flash is Google DeepMind general availability Flash model for coding, agentic execution, spatial reasoning, multimodal understanding, and long-context production workflows. Official DeepMind evidence lists 1M input tokens, 64k output tokens, text, image, video, audio, and PDF inputs, text output, tool use, Gemini API availability, and $1.50 input / $7.50 output pricing per million tokens.
Perplexity AI
Sonar Deep Research
Sonar Deep Research is a large language model from Perplexity AI. Supports up to 128,000 token context window. Achieves 83.0% on MMLU. Available from $2.00/M input tokens.
More Comparisons
Looking for more AI models?
Browse All LLMs