Inference
EN

Claude Opus 5.5 vs Gemini 4 Hydrogen

Change models
Claude Opus 5.5Anthropic
Gemini 4 HydrogenGoogle
IndexAll categories, out of 100.
78.5 ±2.4 best
76.9 ±8.4
GapIndex points below the top.
Top
−1.6 · no clear gap
CoverageIndex weight that has results.
95% best
65%
SpeedOutput tokens per second.
76/s
—
EloLMArena rating and rank.
1513
—
ReleasedMonth of release.
2026-09
2026-11
Input / 1MUS dollars per 1M input tokens.
$4.00
$1.50 best
Output / 1MUS dollars per 1M output tokens.
$20.00
$7.50 best
Batch output / 1MBatch price per 1M output tokens.
$10.00
—
1K chats2K tokens in, 500 out, each.
$18.00
$6.75 best
1K documents20K tokens in, 1K out, each.
$100.00
$37.50 best
1K agent tasks200K tokens in, 20K out, each.
$1,200.00
$450.00 best
ContextMaximum tokens in one request.
1M
1M
AgenticMulti-step tasks with tools.
74.0
78.5 best
CodingCode writing and repair.
81.5 best
79.1
ReasoningLogic problems and puzzles.
80.8
—
MultimodalTasks with images and text.
74.9
80.5 best
KnowledgeFacts and expert knowledge.
79.0 best
74.3
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
59.5
—
MathMath problems.
77.5
—

Benchmarks

11 shared
Only Claude Opus 5.5: 148
Claude Opus 5.5
Gemini 4 Hydrogen
Artificial Analysis Omniscience AccuracyKnowledge
65.5% best3 / 165
50.5%19 / 165
FrontierCode 2.0Coding
45.5% best2 / 28
38.2%7 / 28
Artificial Analysis AutomationBenchAgentic
69.5%7 / 23
76.1% best4 / 23
Artificial Analysis GDP.pdfAgentic
26.2%9 / 22
27.6% best8 / 22
Artificial Analysis Terminal-Bench v4.0Agentic
59.6% best4 / 22
53.5%10 / 22
cursorBench40Coding
57.8% best3 / 21
54.3%5 / 21
Multimodal Knowledge Agent BenchmarkAgentic
53.0%4 / 20
53.4% best3 / 20
CWE-bench v2Agentic
67.1%3 / 14
68.1% best2 / 14
Terminal-Bench-Science 0.1Agentic
58.7% best5 / 12
51.4%9 / 12
OSWorld 2.1 offline subset (82 tasks)Agentic
87.2% best2 / 11
77.6%6 / 11
Chartography without toolsMultimodal
64.4%5 / 10
66.6% best4 / 10