Gemini 4 Argon
Google's Gemini 4 Argon is a flagship reasoning model for complex software engineering, mathematical and scientific problem solving, and multi-step agents. It accepts text, images, audio, video, and PDFs, with tool calling and structured outputs. Its 1,048,576-token context is shared across input, thinking, and output; synchronous output is capped at 65,536 tokens.
- Reasoning
- Vision
- Audio Input
- Video Input
- Native PDF input
- Tool Calling
- Structured Output
Added Oct 13, 2026
Pricing
Auto routing · per 1M tokens- Input
- $2.00
- Output
- $10.00
- Cache read
- $0.10
Specifications
- Context window
- 1M
- Max output
- 65.5K
- Avg output (7d)
- 1.3K tokens
- Longer than 80% of models
Benchmarks
Benchmarks
Sourced from Artificial Analysis.
Intelligence Index
52.6
Agentic work
AutomationBench-AA
Workflow automation with guardrail penalties
77.5%
Better than 99% of models compared
AutomationBench-AA Tasks Completed
Fully completed workflows without guardrail violations
53.1%
Better than 97% of models compared
AA-Briefcase
Agentic knowledge work (Elo)
1490 Elo
Better than 85% of models compared
GDPval-AA v2
Economically valuable tasks (Elo)
1626 Elo
Better than 93% of models compared
Document reasoning
GDP.pdf
Professional PDF reasoning: all-pass rate
21.8%
Better than 76% of models compared
AA-LCR v1.1
Long context reasoning with updated grading
79.7%
Better than 85% of models compared
Reasoning
HLE
Humanity's Last Exam
57.1%
Better than 99% of models compared
CritPt
Research-level physics reasoning
27.1%
Coding
Terminal-Bench v4.0
Practical coding and terminal tasks
57.1%
Better than 97% of models compared
SciCode
Python programming for scientific computing
61.8%
Better than 98% of models compared
Knowledge
AA-Omniscience Accuracy
Proportion of correctly answered questions
49.9%
AA-Omniscience Hallucination Rate
Rate of incorrect answers among non-correct responses
15.1%
Legacy benchmarks
AA-LCR (unversioned / legacy)
Long context reasoning evaluation
79.7%
Better than 84% of models compared
GDPval-AA (unversioned / legacy)
Economically valuable tasks
56.2%
Last updated Oct 14, 2026
Artificial AnalysisProviders
Auto routing is available for this model. Explicit provider selection is not available.
Loading provider options…