Claude Haiku 5.5 vs Tencent Hy4

Compare up to 4 models side by side.


Capability
Capability	160	—
Human preference	1401	—
Priceper 1M tokens
Input	$0.11–$0.121↓	$0.438list $0.876-50%↑ ×4.0
Output	$0.55–$0.605↓	$1.314list $2.628-50%↑ ×2.4
Cache read	$0.011–$0.0121↓	$0.0219list $0.0438-50%↑ ×2.0
Web searchper request	$0.011	—
Limits and modalities
Context	1M↓ 5%	1M↑
Max output	128K↑	64K↓ 50%
Modalities	InputTextImagePDFOutputText	InputTextImageOutputText
Availability
Providers	Anthropic, Amazon Bedrock, Anthropic (AWS)	Tencent Cloud TokenHub
30-day uptime	—	—
Measured, last 30 days
Avg latency	—	—
Effective price	—	—
Cache hit rate	—	—
About
Released	2026-10-07	2026-11-04
Knowledge cutoff	2026-06	—

Capability scores are from Epoch AI’s Epoch Capabilities Index (CC BY 4.0, as of 2026-11-02), which combines dozens of benchmarks with GPT-5 set at 150. epoch.ai/eci

Human preference ratings are from the LMArena Text Arena (CC BY 4.0, as of 2026-10-28): Elo ratings from blind votes, on a different scale from capability scores. lmarena.ai/leaderboard/text