Hy3-preview: benchmark scores, price and ranking

Hy3-preview from tencent ranks #110 of 743 LLMs on the LLMs Tiger aggregate benchmark score (51.6), based on 1 benchmark results captured up to 2026-10-13.

Organisation: tencent
Type: Open-weight / local model
Open weights: Yes
Cheapest API price: $0.083 in / $0.330 out per million tokens (via openrouter, observed 2026-10-13T20:47Z)
Aggregate rank: #110 of 743 (score 51.6)

Across the 1 benchmark where at least ten models are scored, Hy3-preview is in the top 10% on 0 and in the top half on 1.

Its cheapest listed input price, $0.083 per million tokens, is lower than that of 85% of the 427 models with a listed price.

Hy3-preview benchmark results

Hy3-preview benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
GPQA Diamond	87.2%	#59 of 275	HuggingFace GPQA Diamond leaderboard	2026-10-13

About Hy3-preview

Hy3-preview is an open-weight model that can run on your own hardware from tencent. The cheapest API listing we track charges $0.083 per million input tokens and $0.330 per million output tokens, through openrouter. We hold 1 benchmark result for it across 1 area, and none is in the top tenth of its board.

Where Hy3-preview stands

On our aggregate score, Hy3-preview is #110 of 743 ranked models, in the top quarter. Its score is 51.6; the scale is centred on 50, so it sits 1.6 points above the typical model. That beats 85% of the ranked field. Among the 5 ranked tencent models it is #3. The next one up is Hy4-preview at 51.9. The next one down is Hunyuan-A13B-Instruct at 49.5.

Hy3-preview, benchmark by benchmark

GPQA Diamond puts Hy3-preview at 87.2%, #59 of 275. That beats 79% of the models tested, but 8.6 points short of GPT-6 Astra (max) (95.8%). The results either side are Qwen 3.6 Max (Preview) (87.4%) and grok-4-0709 (87.0%). It covers expert-level science questions. The board median is 71.7%, so Hy3-preview is 15.5 points above it. Source: HuggingFace GPQA Diamond leaderboard, captured 2026-10-13.

Price and value

The median input price across the 427 priced models is $0.500 per million tokens, so Hy3-preview costs less than the typical listing. Models that score within 1.5 points or better and cost less: DeepSeek V4 Flash 0731 (max) ($0.016, 53.1), Qwen3.7 Flash ($0.030, 50.5), GLM-5.2 (high) ($0.060, 51.6). Similar scores cost more on Step-3.5-Flash ($0.100) and Claude Haiku 5.5 (max) ($0.100).

Reading these numbers

The 1 score on this page comes from 1 source: HuggingFace GPQA Diamond leaderboard (1). No official leaderboard supplied any of them, so treat them as third-party measurements. With 1 results the aggregate rank can move a lot when one new score arrives.

Models ranked nearby

#107 o1-2024-12-17 (51.6)
#108 DeepSeek v4 Pro (max) (51.6)
#109 GLM-5.2 (high) (51.6)
#111 LongCat-Flash-Chat (51.6)
#112 Darwin-27B-Opus (51.5)
#113 Muse Spark 1.2 (xhigh) (51.5)