Compare AI models

Specifications
Context window
1M1M
API pricing
Input price
$0.10$2.00
Output price
$0.50$10.00
Cached input price
$0.01$0.10
Cheapest input
$0.10Amazon Bedrock$1.00Google AI Studio
Cheapest output
$0.50Amazon Bedrock$5.00Google AI Studio
$45.00Anthropic list price$900Google list price
Benchmarks
BullshitBench v2
57%66%
CursorBench 4.0
48.4%52.9%
FrontierSWE V2
43.8%55%
Terminal-Bench 4.0
39.2%57.4%
Terminal-Bench-Science 0.1
20.6%57.6%
ARC-AGI-2
61.25%91.25%
Chartography · no tools
46.4%71.6%
threejseval
14241702
Overview
CompanyAnthropicGoogle
Release dateOct 7 2026Sep 30 2026
AccessClosedClosed
Model detailsView modelView model

Frequently asked questions

Gemini 4 Argon leads Claude Haiku 5.5 on 8 of the 8 benchmarks they both report. Claude Haiku 5.5 is cheaper on both input and output: $0.10 vs $2.00 per million input tokens, and $0.50 vs $10.00 per million output tokens. Figures are base-tier rates. Gemini 4 Argon shipped 7 days before Claude Haiku 5.5, so benchmark comparisons should account for the intervening progress.

Context windows are 1M (Claude Haiku 5.5) vs 1M (Gemini 4 Argon).

On BullshitBench v2, Gemini 4 Argon leads at 66% vs Claude Haiku 5.5 at 57%. On CursorBench 4.0, Gemini 4 Argon leads at 52.9% vs Claude Haiku 5.5 at 48.4%. On FrontierSWE V2, Gemini 4 Argon leads at 55% vs Claude Haiku 5.5 at 43.8%. On Terminal-Bench 4.0, Gemini 4 Argon leads at 57.4% vs Claude Haiku 5.5 at 39.2%. On Terminal-Bench-Science 0.1, Gemini 4 Argon leads at 57.6% vs Claude Haiku 5.5 at 20.6%. On ARC-AGI-2, Gemini 4 Argon leads at 91.25% vs Claude Haiku 5.5 at 61.25%. On Chartography · no tools, Gemini 4 Argon leads at 71.6% vs Claude Haiku 5.5 at 46.4%. On threejseval, Gemini 4 Argon leads at 1702 vs Claude Haiku 5.5 at 1424.