Claude Sonnet 4.6 vs Gemini 3.1 Pro

Benchmarks, API pricing and specs, head to head. Data updated 2026-06-10.

Claude Sonnet 4.6

Anthropic · Apr 2026

Anthropic's mid-tier workhorse: near-Opus capability at $3/$15, and the leader on the Finance Agent benchmark.

Gemini 3.1 Pro

Google · Mar 2026

Google's flagship with the industry's largest context window — 2M tokens — and frontier multimodal reasoning at $2/$12.

The verdict

Claude Sonnet 4.6 wins 1 of the 1 benchmarks these models share, against 0 for Gemini 3.1 Pro. Gemini 3.1 Pro is about 1.3x cheaper per blended million tokens (3:1 input:output mix). Gemini 3.1 Pro also takes 2M of context versus 1M for Claude Sonnet 4.6.

Benchmark head-to-head 10

SWE-bench Verified
~79.6%~78%
Claude Sonnet 4.6Gemini 3.1 Pro

Specs & pricing

Claude Sonnet 4.6Gemini 3.1 Pro
modhub Index
Input price / 1M$3$2
Output price / 1M$15$12
Context window1M2M
Max output64K66K
Open weightsnono
Reasoning modelyesyes
Multimodal inputtext, imagetext, image, audio, video
Knowledge cutoffJan 2026Nov 2025
ReleasedApr 2026Mar 2026
Example monthly cost*$52.50$38.00

* 10M input + 1.5M output tokens per month at list prices, no caching. Green = better value on that row.

Frequently asked questions

Which is better, Claude Sonnet 4.6 or Gemini 3.1 Pro?
Claude Sonnet 4.6 wins 1 of the 1 benchmarks these models share, against 0 for Gemini 3.1 Pro. Gemini 3.1 Pro is about 1.3x cheaper per blended million tokens (3:1 input:output mix). Gemini 3.1 Pro also takes 2M of context versus 1M for Claude Sonnet 4.6.
Which is cheaper, Claude Sonnet 4.6 or Gemini 3.1 Pro?
Claude Sonnet 4.6 costs $3/$15 per million input/output tokens, while Gemini 3.1 Pro costs $2/$12. For a typical workload of 10M input and 1.5M output tokens per month, that's $52.50 versus $38.00.
Which model is better for coding, Claude Sonnet 4.6 or Gemini 3.1 Pro?
On SWE-bench Verified — the standard agentic-coding benchmark — Claude Sonnet 4.6 scores ~79.6% versus ~78% for Gemini 3.1 Pro, making Claude Sonnet 4.6 the stronger pick for coding agents.

More comparisons