Gemini 3.6 Flash vs Claude Sonnet 4.6
Side-by-side comparison of API pricing, specs, benchmarks, and capabilities
| Specification |
Gemini 3.6 FlashIntro price
by Google
|
by Anthropic
|
|---|---|---|
| Overview | ||
| Status | Current flagship | Current mid tier |
| Released | ||
| Pricing per million tokens | ||
| Input | $0.750/Mtok | $3.00/Mtok |
| Output | $3.75/Mtok | $15.00/Mtok |
| Blended avg | $1.50/Mtok | $6.00/Mtok |
| Cached input | $0.075/Mtok | $0.300/Mtok |
| Price basis | Google's Gemini API pricing page prints a dated, two-phase schedule for this model: $0.75 per 1M input tokens through December 31, 2026, rising to $1.50 on January 1, 2027, and $3.75 per 1M output tokens through December 31, 2026, rising to $7.50. Context caching follows the same dates ($0.075, then $0.15). We publish the rate you would pay today and re-verify it before the changeover date. Checked 2026-08-13 against ai.google.dev/gemini-api/docs/pricing; the same page listed a flat $1.50/$7.50 on 2026-08-09, so this is a genuine 50% price cut rather than a correction to an earlier mistake. | — |
| Specifications | ||
| Context window | 1M tokens | 1M tokens |
| Parameters | Proprietary | Proprietary |
| Speed (TPS) | — | — |
| Modalities | ||
| Input |
textimageaudiovideo
|
textimage
|
| Benchmarks sources: Artificial Analysis, Google DeepMind Gemini 3.6 Flash model card, LMSYS Chatbot Arena (UC Berkeley) / Vellum LLM Leaderboard | ||
| Terminal-Bench 2.1 | 78 vendor | 59.1 |
| OSWorld-Verified | 83 vendor | 78.5 |
| Avg benchmark score | — | 70.4 |
| Perf / dollar | — | 11.7 |
| GPQA Diamond | — | 89.9 |
| LiveCodeBench | — | 72.4 |
| MCP Atlas | — | 69.5 |
| BrowseComp | — | 76.2 |
| Humanity's Last Exam | — | 19.1 |
| ARC-AGI 2 | — | 58.3 |
| AIME 2025 | — | 83 |
| MMMLU | — | 89.3 |
| HumanEval | — | 79.6 |
| AutoBench | — | 5.3 |
| Independent composite scores each on its own scale — not part of the average above | ||
| AA Intelligence Index | 51.6 | 48.4 |
| AA Agentic Index | 40.5 | 42.1 |
| AA-Omniscience Index (−100–100) | 22.1 | 12.2 |
| LMSYS Chatbot Arena | 1480.2 ELO | 1472.4 ELO |
| GDPval-AA v2 | 1422.3 ELO | — |
| Providers | ||
| Available from |
Google — $0.750/$3.75/Mtok
|
Anthropic — $3.00/$15.00/Mtok
|
Cost at scale
1M tokens · 50/50 input/output| Volume | Gemini 3.6 Flash | Claude Sonnet 4.6 | Savings |
|---|---|---|---|
| 1M tokens | $1.5 | $6 | $4.5 (75%) |
| 10M tokens | $15 | $60 | $45 (75%) |
| 100M tokens | $150 | $600 | $450 (75%) |
| 1000M tokens | $1500 | $6000 | $4500 (75%) |
When to pick which
distilled from the pricing and spec data above- cost dominates: $1.50/Mtok blended vs $6.00 — 75% less on the same 50/50 token mix
- this pairing gives Claude Sonnet 4.6 no edge on price, context, cache discount, or measured performance — pick it only for qualitative fit (ecosystem, compliance, model behavior on your prompts)
Price movement
recorded changes since we began tracking each model · last checked| Model | Changes | Latest move | Old → new $/M |
|---|---|---|---|
| Gemini 3.6 Flash | 1 | price cut on | $1.5/$7.5 → $0.75/$3.75/Mtok |
| Claude Sonnet 4.6 | 0 | — no move | Stable since we began tracking — no change recorded |
The cheaper option here — Gemini 3.6 Flash costs $1.50/Mtok blended on Google.
Summary
Gemini 3.6 Flash by Google costs $0.750/Mtok input and $3.75/Mtok output, with a 1M-token context window. It supports text, image, audio, video input.
Claude Sonnet 4.6 by Anthropic costs $3.00/Mtok input and $15.00/Mtok output, with a 1M-token context window. It supports text, image input.
On a blended cost basis, Gemini 3.6 Flash is 75% cheaper than Claude Sonnet 4.6.
The two aren't directly comparable on average benchmark score: Claude Sonnet 4.6 has published per-benchmark results, while Gemini 3.6 Flash does not yet — it is measured today on independent composites (see the table above).
Note: Pricing is per million tokens. Actual costs vary with usage patterns, prompt caching, and batch discounts. Always verify against official provider pricing pages.