Tencent

Hy3

AI model by Tencent. Real-time pricing and benchmark data.

Pricing (per 1M tokens)

Input$0.14
Output$0.56
Blended (3:1)$0.24

Source: Artificial Analysis

Performance

Output Speed71 tok/s
Time to First Token1757ms

Median values from Artificial Analysis

Editorial Analysis

Profile. Hy3 is a budget-tier option from Tencent at $0.56/M output tokens. Despite the low price, it scores competitively on GPQA Diamond (89.7%) and Coding Index (58.8) — making it a candidate for high-volume production workloads where unit cost dominates.

Latency. On latency, Hy3 reports 71 tok/s output speed and 1757ms time-to-first-token. Output speed is on the slower end; for real-time chat consider pairing with a streaming front-end or routing latency-sensitive requests to a faster model.

Cost model. For a workload of 1M input tokens + 100K output tokens, Hy3 costs $0.19 per million input-equivalent requests. Multiply by your monthly request volume to project the bill, and compare against the speed/correctness profile above to decide if the trade-off is worth it for your workload.

Editorial summary generated from public pricing and benchmark data (Artificial Analysis API). All numbers are sourced — see the Artificial Analysis methodology for benchmark definitions. Last refreshed 2026-08-03.

Compare with similar models

ModelInputOutputSpeed
Hy3Current
$0.14$0.5671 tok/s
Llama 3.1 Instruct 70B
$0.56$0.560 tok/s
Ring-flash-2.0
$0.14$0.570 tok/s
Ling-flash-2.0
$0.14$0.570 tok/s
Seed-OSS-36B-Instruct
$0.21$0.570 tok/s
gpt-oss-120b (low)
$0.15$0.59275 tok/s

Example Costs

Single Request
$0.0004
1.0K in / 500 out
1K Requests/day
$0.4145
1.0M in / 500.0K out
10K Requests/day
$4.15
10.0M in / 5.0M out