Compare/Hy3-preview (Non-reasoning) vs o3

Hy3-preview (Non-reasoning)vso3

Side-by-side comparison of pricing, 12 benchmarks, and generation speed.

Tencent

Hy3-preview (Non-reasoning)

Input
$0/M
Output
$0/M
Speed
89 tok/s
TTFT
2.35s
OpenAI

o3

Input
$2/M
Output
$8/M
Speed
74 tok/s
TTFT
8.58s

Winner by Category

Cheaper
Hy3-preview (Non-reasoning)
Faster (tok/s)
Hy3-preview (Non-reasoning)
Lower Latency
Hy3-preview (Non-reasoning)
Benchmarks (0-12)
o3

Pricing Comparison

MetricHy3-preview (Non-reasoning)o3
Input ($/M tokens)$0$2
Output ($/M tokens)$0$8
Cost for 1M input + 100K output tokens:
Hy3-preview (Non-reasoning)$0.00
o3$2.80

Speed Comparison

Output Speed (tokens/s) — higher is better
Hy3-preview (Non-reasoning)
89 tok/s
o3
74 tok/s
Time to First Token (seconds) — lower is better
Hy3-preview (Non-reasoning)
2.35s
o3
8.58s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
33.738.4
Coding Index
34.338.4
Math Index
88.3
GPQA Diamond
73.2%82.7%
MMLU-Pro
85.3%
LiveCodeBench
80.8%
AIME 2025
88.3%
MATH-500
99.2%
Humanity's Last Exam
6.3%20.0%
SciCode
39.4%41.0%
IFBench
48.0%71.4%
TerminalBench
31.8%37.1%
Hy3-preview (Non-reasoning)0 wins
12 winso3

Frequently Asked Questions

Which is cheaper, Hy3-preview (Non-reasoning) or o3?

Hy3-preview (Non-reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.00/M tokens vs $3.50/M for o3.

Which model performs better on benchmarks?

o3 wins 12 out of 12 benchmarks compared to 0 for Hy3-preview (Non-reasoning). See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

Hy3-preview (Non-reasoning) generates tokens faster at 89 tok/s vs 74 tok/s. Hy3-preview (Non-reasoning) also has lower time-to-first-token (2.35s vs 8.58s).

When should I use Hy3-preview (Non-reasoning) vs o3?

Choose based on your priorities: Hy3-preview (Non-reasoning) for lower cost, o3 for stronger benchmark performance, and Hy3-preview (Non-reasoning) for faster generation. For latency-sensitive apps, check the TTFT comparison above.