Inference
EN

Claude Opus 5.5 vs Hy4

Change models
Claude Opus 5.5Anthropic
Hy4Tencent
IndexAll categories, out of 100.
78.6 ±2.4 best
72.0 ±7.8
GapIndex points below the top.
Top
−6.6 · no clear gap
CoverageIndex weight that has results.
95% best
65%
SpeedOutput tokens per second.
76/s best
52/s
EloLMArena rating and rank.
1513
—
ReleasedMonth of release.
2026-09
2026-11
Input / 1MUS dollars per 1M input tokens.
$4.00
$0.417 best
Output / 1MUS dollars per 1M output tokens.
$20.00
$1.25 best
Batch output / 1MBatch price per 1M output tokens.
$10.00
—
1K chats2K tokens in, 500 out, each.
$18.00
$1.46 best
1K documents20K tokens in, 1K out, each.
$100.00
$9.59 best
1K agent tasks200K tokens in, 20K out, each.
$1,200.00
$108.41 best
ContextMaximum tokens in one request.
1M
1.05M best
AgenticMulti-step tasks with tools.
74.1
74.5 best
CodingCode writing and repair.
81.8 best
68.4
ReasoningLogic problems and puzzles.
80.8
—
MultimodalTasks with images and text.
74.1 best
57.4
KnowledgeFacts and expert knowledge.
79.2
83.7 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
59.6
—
MathMath problems.
77.7
—

Benchmarks

14 shared
Only Claude Opus 5.5: 143Only Hy4: 7
Claude Opus 5.5
Hy4
Vibe Code Bench v1.1Coding
90.3% best6 / 123
81.0%30 / 123
Legal Research BenchLegal
50.5% best7 / 90
41.3%30 / 90
Tax Agent BenchFinance
44.5% best6 / 82
34.4%17 / 82
Terminal-Bench 4.0Agentic
65.2% best1 / 61
39.4%15 / 61
DeepSWEAgentic
74.2%10 / 59
74.9% best8 / 59
Agent Arena steerabilityAgentic
18.6 best3 / 54
8.814 / 54
Agent Arena task outcomeAgentic
20.7 best2 / 54
6.712 / 54
Humanity's Last Exam with toolsAgentic
67.7% best2 / 36
59.1%10 / 36
HLAB-Vals-1.1Legal
9.2%22 / 35
12.5% best17 / 35
AutomationBenchAgentic
40.0%20 / 32
56.2% best4 / 32
FrontierCode 2.0Coding
45.5% best2 / 27
31.0%9 / 27
ProgramBench: Can Language Models Rebuild Programs From Scratch?Coding
91.2% best2 / 19
30.0%12 / 19
OSWorld 2.1 (all 108 tasks)Agentic
81.8% best1 / 11
78.9%5 / 11
Terminal-Bench-Science 0.1Agentic
58.7% best5 / 11
19.8%11 / 11