Inference
EN

Gemini 4 Carbon vs Hy4

Change models
Gemini 4 CarbonGoogle
Hy4Tencent
IndexAll categories, out of 100.
82.7 ±5.1 best
72.0 ±7.8
GapIndex points below the top.
Top
−10.7
CoverageIndex weight that has results.
80% best
65%
SpeedOutput tokens per second.
31/s
52/s best
EloLMArena rating and rank.
1538
—
ReleasedMonth of release.
2026-10
2026-11
Input / 1MUS dollars per 1M input tokens.
$2.00
$0.417 best
Output / 1MUS dollars per 1M output tokens.
$10.00
$1.25 best
Batch output / 1MBatch price per 1M output tokens.
$5.00
—
1K chats2K tokens in, 500 out, each.
$9.00
$1.46 best
1K documents20K tokens in, 1K out, each.
$50.00
$9.59 best
1K agent tasks200K tokens in, 20K out, each.
$600.00
$108.41 best
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
83.2 best
74.5
CodingCode writing and repair.
80.5 best
68.4
ReasoningLogic problems and puzzles.
83.9
—
MultimodalTasks with images and text.
—
57.4
KnowledgeFacts and expert knowledge.
79.5
83.7 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
—
—
MathMath problems.
77.2
—

Benchmarks

9 shared
Only Gemini 4 Carbon: 30Only Hy4: 12
Gemini 4 Carbon
Hy4
Vibe Code Bench v1.1Coding
89.1% best9 / 123
81.0%30 / 123
Legal Research BenchLegal
55.8% best1 / 90
41.3%30 / 90
Tax Agent BenchFinance
51.8% best1 / 82
34.4%17 / 82
Terminal-Bench 4.0Agentic
63.1% best4 / 61
39.4%15 / 61
Agent Arena steerabilityAgentic
19.1 best2 / 54
8.814 / 54
Agent Arena task outcomeAgentic
18.4 best4 / 54
6.712 / 54
HLAB-Vals-1.1Legal
17.5% best8 / 35
12.5%17 / 35
AutomationBenchAgentic
53.9%5 / 32
56.2% best4 / 32
FrontierCode 2.0Coding
40.0% best5 / 27
31.0%9 / 27