Compare/Hy3-preview (Reasoning) vs GPT-5.4 mini (medium)

Hy3-preview (Reasoning)vsGPT-5.4 mini (medium)

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

Tencent

Hy3-preview (Reasoning)

Input
$0/M
Output
$0/M
Speed
86 tok/s
TTFT
2.40s
OpenAI

GPT-5.4 mini (medium)

Input
$0.75/M
Output
$4.5/M
Speed
161 tok/s
TTFT
3.79s

Winner by Category

Cheaper
Hy3-preview (Reasoning)
Faster (tok/s)
GPT-5.4 mini (medium)
Lower Latency
Hy3-preview (Reasoning)
Benchmarks (3-3)
Tie

Pricing Comparison

MetricHy3-preview (Reasoning)GPT-5.4 mini (medium)
Input ($/M tokens)$0$0.75
Output ($/M tokens)$0$4.5
Cost for 1M input + 100K output tokens:
Hy3-preview (Reasoning)$0.00
GPT-5.4 mini (medium)$1.20

Speed Comparison

Output Speed (tokens/s) — higher is better
Hy3-preview (Reasoning)
86 tok/s
GPT-5.4 mini (medium)
161 tok/s
Time to First Token (seconds) — lower is better
Hy3-preview (Reasoning)
2.40s
GPT-5.4 mini (medium)
3.79s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
41.937.7
Coding Index
36.537.5
Math Index
GPQA Diamond
86.7%82.3%
MMLU-Pro
LiveCodeBench
AIME 2025
MATH-500
Humanity's Last Exam
25.5%17.1%
SciCode
41.2%44.2%
IFBench
63.1%64.8%
TerminalBench
34.1%34.1%
Hy3-preview (Reasoning)3 wins
3 winsGPT-5.4 mini (medium)

Frequently Asked Questions

Which is cheaper, Hy3-preview (Reasoning) or GPT-5.4 mini (medium)?

Hy3-preview (Reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.00/M tokens vs $1.69/M for GPT-5.4 mini (medium).

Which model performs better on benchmarks?

It's a tie — both models win 3 benchmarks each across 12 evaluated categories. See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

GPT-5.4 mini (medium) generates tokens faster at 161 tok/s vs 86 tok/s. Hy3-preview (Reasoning) also has lower time-to-first-token (2.40s vs 3.79s).

When should I use Hy3-preview (Reasoning) vs GPT-5.4 mini (medium)?

Choose based on your priorities: Hy3-preview (Reasoning) for lower cost, both perform similarly on benchmarks, and GPT-5.4 mini (medium) for faster generation. For latency-sensitive apps, check the TTFT comparison above.