Gemini 4 Argon vs Gemini 4 Carbon
Wins 0 of 7 areas
—
Wins 5 of 7 areas
Coding · Agents · Reasoning · Facts · Math
Gemini 4 Carbon is the stronger all-rounder.
Scores updated · 28 tests both models report · How we compare
Where each one wins
Tests won in each of the seven areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- MathCompetition and research-level math02Gemini 4 Carbon2 of 2 tests
- ReasoningHard problems that need careful thinking01Gemini 4 Carbon1 of 3 tests · 2 ties
- CodingWriting and fixing software01Gemini 4 Carbon1 of 1 test
- AgentsCarrying out multi-step tasks on its own01Gemini 4 Carbon1 of 1 test
- FactsGetting facts right instead of making them up01Gemini 4 Carbon1 of 1 test
- Images and chartsUnderstanding pictures, charts and video00Even0 each · 1 tie
- Long documentsFinding answers in very long texts00Even0 each · 1 tie
Coding, agents, facts, images and charts and long documents rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Gemini 4 Carbon pulls ahead
- Solves new puzzle games it has never seenARC-AGI-3+17.6points ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+17points ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+8.1points ahead
Where Gemini 4 Argon pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 28 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemini 4 Carbon
- LMArena · WebDevGemini 4 Carbon by 89 rating points16791768+89 rating
ReasoningGemini 4 Carbon
- ARC-AGI-3Gemini 4 Carbon by 17.646.363.9+17.6
- SimpleBenchtie91.792.6tie
- ARC-AGI-2tie91.392.1tie
MathGemini 4 Carbon
- FrontierMath Tier 4Gemini 4 Carbon by 177895.1+17
- FrontierMath Tiers 1-3 (v2)Gemini 4 Carbon by 4.58690.5+4.5
Other results18 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- CritPt v2 HardGemini 4 Carbon by 16.552.569+16.5
- HLE-DiamondGemini 4 Carbon by 12.456.769.1+12.4
- Science and math (Terminal-Bench-Science 0.1 (AA))Gemini 4 Carbon by 11.952.964.8+11.9
- Terminal-Bench-Science 0.1Gemini 4 Carbon by 11.952.964.8+11.9
- HiL-Bench (Tools-allowed)Gemini 4 Argon by 11.76250.3+11.7
- General capabilities (AA Index v5.0)Gemini 4 Carbon by 95059+9
- Computer use (Agents' Last Exam v1.1)Gemini 4 Carbon by 8.46977.4+8.4
- Agentic coding (Terminal-Bench 4.0 (AA))Gemini 4 Carbon by 8.157.165.2+8.1
- AA IntelligenceGemini 4 Carbon by 7.452.660+7.4
- Computer use (OSWorld-2.1)Gemini 4 Carbon by 6.874.781.5+6.8
- GDP (Surge AI)Gemini 4 Carbon by 6.621.828.4+6.6
- RiemannBenchGemini 4 Carbon by 47680+4
- Science and math (RiemannBench)Gemini 4 Argon by 47672+4
- Cybersecurity (CWE-bench)Gemini 4 Carbon by 36871+3
- Knowledge work (AutomationBench)Gemini 4 Carbon by 2.651.353.9+2.6
- Vals IndexGemini 4 Carbon by 1.568.970.4+1.5
- General capabilities (Vals Index v2.2)Gemini 4 Carbon by 1.269.270.4+1.2
- ARC-AGI-1Gemini 4 Argon by 198.597.5+1
Questions people ask
Which is better, Gemini 4 Argon or Gemini 4 Carbon?
Gemini 4 Carbon wins five of the seven areas where both have results: coding, agents, reasoning, facts and math. Gemini 4 Argon wins none. They are level on images and charts and long documents.
Which is better for coding?
Gemini 4 Carbon. It wins the one coding test both models report.
Which is cheaper?
Gemini 4 Argon costs $2.00 per million input tokens and $10.00 per million output tokens; Gemini 4 Carbon costs $2.00 and $10.00. For a million tokens read plus a million written, they cost about the same.
How do you compare the two?
We use the 28 benchmark tests both models have published scores on. The verdict counts the 10 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 18 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.