Atlas

Compare Models

Gemini 4 Carbon vs Hy4

Gemini 4 Carbon leads by 5.8 capability points (≈95% interval +3.1 to +8.6): a clear lead over Hy4 · P(higher) ≈ 100%

Wins 17 of 17 shared benchmarks (0 lost)

Capability vs. relative cost
Gemini 4 CarbonHy4Closed weightsOpen weights

Hy4 has no setting with a relative cost yet.

Benchmark wins17 – 0 · 17 shared

Comparing matching setups only (155 benchmarks excluded) · suites counted once
← Gemini 4 CarbonHy4 →
Data: Surge AI, Mercor, Collinear AI and 4 othersatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Hy4 minus Gemini 4 Carbon.

Chartography (November 2026 rescoring)Perception & media › Documents & charts85.549.2−36.3
RiemannBench (November 2026 rescoring)Math80.051.2−28.8
DAYJOB: Healthcare (October 2026 rescoring)Health & medicine35.27.2−28.0
HANDBOOK.md Agents v1.1Tool & computer use › Customer service & operations49.823.7−26.1
GDP.xlsxProfessional › General knowledge work38.917.7−21.2
DAYJOB: Finance (October 2026 rescoring)Professional › Finance, accounting & tax23.89.2−14.6
GDP.pdf (November 2026 rescoring)Professional › General knowledge work56.443.6−12.8
FoodTruck BenchBusiness simulations7721867598−12.3
EnterpriseBench: CoreCraftTool & computer use › Customer service & operations80.568.2−12.3
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Professional › Legal23.013.6−9.4
τ³-Banking (Mercor run, bm25_grep retrieval)Tool & computer use › Customer service & operations55.748.1−7.6
DAXBench v2Software › Data & ML93.687.2−6.4
APEX-Agents 1.1Professional › General knowledge work90.485.1−5.3
Agents' Last Exam v1.1 (ALE v1.1)Professional › General knowledge work45.941.0−4.9
LLM Debate BenchmarkLanguage & writing › Language & wordplay16761647−4.2
ComplexConstraints (November 2026 rescoring)Language & writing › Instruction following70.966.8−4.1
CWE-bench v2 Pass@1Software › Cybersecurity67.364.1−3.2
Data: Surge AI, Mercor, Collinear AI and 4 othersatlas.village.dev
Win counts use each benchmark’s declared score direction. Represented suite slices and the derived Capability score are excluded from headline totals. Missing or incompatible setups are excluded, never counted as losses.
Loading Atlas data…