Models / Hy3 Preview

Tencent

Hy3 Preview

Hy3 Preview is strongest at reasoning and knowledge; worth a look for simple, high-volume work where cost matters most. The cheapest model of similar intelligence is DeepSeek-V4-Flash (0423), at $0.15 per million tokens against $0.29.

Superseded. Hy4 Preview is Tencent's newer model in this line, at $1.25 blended: see Hy4 Preview.

Results 16 Nov 2026 · 27 results · 3 sources

Price per million tokens
$0.18 in · $0.60 out
Typical host on OpenRouter, 10 Nov 2026 · compare prices
Speed
Not measured
Intelligence Index
22.7
Artificial Analysis, at thinking reasoning; 17.0–22.7 across 2 settings
Spring Prompt overall
Not ranked yet
needs our benchmarks and two groups of results
Developer
Tencent, based in China
Weights published
Released
23 Apr 2026 (Artificial Analysis)
262,144 tokens of context

Where it stands among the models you could choose

Available models grouped by price, intelligence, speed and developer location, plus models with published weights. Retired models are excluded.

AmongIntelligence IndexSpring Prompt overallPrice, blendedSpeed
Models available today 87th of 185Claude Opus 5.5 – $0.29 blended, 26th percentilecheapest: Mistral Nemo –
Models priced $0.25–1 per million tokens 18th of 61MiMo-V2.6-Pro – sets this group –
Models with similar intelligence sets this group – $0.29 blended, 11th percentilecheapest: DeepSeek-V4-Flash (0423) –
Models from developers based in China 34th of 75MiMo-V2.6-Pro – $0.29 blended, 21st percentilecheapest: Qwen3.5-9B –
Open-weights models 34th of 104MiMo-V2.6-Pro – $0.29 blended, 36th percentilecheapest: Mistral Nemo –

The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.

How good is it, and for what?

Each use case ranks the available models its benchmarks measured, then those priced $0.25–1 per million tokens. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.

Product listingsTurning a sparse product feed and photos into listings that can go live Not measured
  • CatalogBench: has not measured this model.
Decks from an analysisTurning a finished analysis into a deck you could present as it is Not measured
  • DeckBench: has not measured this model.
User surveysPlanning a user survey and reading its results without being misled Not measured
  • SurveyBench: has not measured this model.
Marketing planningPlanning a year of ad spend without overspending Not measured
  • ROASBench: has not measured this model.
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls PendingReleased after these sources last updated
  • tau2-bench: has not measured this model.
  • Berkeley Function Calling Leaderboard (BFCL) V4: pending. Its latest results are dated 16 Dec 2025, before this model was released.
  • OpenHands Index: has not measured this model.
  • Microsoft STATE-Bench: has not measured this model.
  • Artificial Analysis: has not measured this model.
  • Vending-Bench 2: has not measured this model.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects Not measured
  • APEX-Agents: has not measured this model.
  • GDP.pdf: has not measured this model.
  • Remote Labor Index: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge 86th of 187 available · 16th of 61 at its price
BenchmarkHy3 PreviewRank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 22.7 87th of 185 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 27.8% 84th of 186 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 86.7% 61st of 168 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 89th of 169 available · 19th of 58 at its price
BenchmarkHy3 PreviewRank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,409 92nd of 153 Gemini 4 Argon 1,525
UGI LeaderboardUGI: writing score · reported 51.1 56th of 124 Gemini 3.8 Flash 78.6
Sticking to the factsSummarising without inventing things, and factual answers 121st of 139 available · 31st of 40 at its price
BenchmarkHy3 PreviewRank among availableBest available
Arena (formerly LMArena)Arena Text factuality: overall · reported 1,412 91st of 109 Gemini 4 Argon 1,511
  • Vectara Hallucination Leaderboard: has not measured this model.
  • SimpleQA Verified (Epoch AI): has not measured this model.
Speed under pressureGood decisions against a real clock (fast chess) Not measured
  • BulletBench: has not measured this model.

Nearest alternatives

Against Hy3 Preview at $0.29 blended, Intelligence Index 22.7. "Similar intelligence" means within 4 points or better.

Where to run it

1 host on OpenRouter at the standard service tier, with a median of $0.28 blended per million tokens; quantisation where the host states it: BF16. Cheapest first.

HostQuantisationInputOutputBlended
GMICloudBF16$0.18$0.60$0.29

Prices per million tokens from OpenRouter's public endpoint list, checked 10 Nov 2026.

Against the alternatives

The models you would most likely weigh it against: the leaders of the groups above.

Scroll sideways to see every alternative.

Hy3 PreviewMiMo-V2.6-Pro
most intelligent at $0.25–1 per million tokens
DeepSeek-V4-Flash (0423)
cheapest with similar intelligence
Claude Opus 5.5
most intelligent available today
Price per million tokens $0.29 $0.54$0.15$8.00
Tokens a second – 43–96
Intelligence Index 22.7 46.324.457.6
Spring Prompt overall – ––82
Rank among available models, by use case
Product listings – ––13th
Decks from an analysis – ––7th
User surveys – ––4th
Marketing planning – ––9th
Agents and tool use pending 23rd59th3rd
Professional work – ––3rd
Reasoning and knowledge 86th of 187 13th69th1st
Writing 89th of 169 23rd63rd5th
Sticking to the facts 121st of 139 10th81st3rd
Speed under pressure – –––

Shaded figures are better than Hy3 Preview's. A dash means no result.

Where it has been measured

18th most intelligent of the 61 models priced $0.25–1 per million tokens. Among models available today, in the bottom quarter for sticking to the facts. Results for agents and tool use are pending: those sources last updated before this model was released.

Artificial Analysis lists it but has not published a speed for it yet

Every result

27 published results from 3 sources, each in the source's own units, with the configuration that produced it.

Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 7 results
MeasureValueRankConfigurationDated
Arena Text factuality: overallArena rating, higher is better 1,412
1,405–1,419
143 of 182 Hy3 Preview 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,411
1,396–1,427
141 of 406 Hy3 Preview 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,353
1,334–1,372
172 of 411 Hy3 Preview 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,446
1,424–1,468
122 of 363 Hy3 Preview 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,396
1,384–1,409
151 of 413 Hy3 Preview 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,409
1,401–1,416
149 of 413 Hy3 Preview 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,377
1,362–1,392
163 of 412 Hy3 Preview 2 Oct 2026

Not shown: Not an error rate: claims that cannot be checked on the web are skipped, and preference still carries most of the weight.

Artificial Analysis · reported by Artificial Analysis · 14 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 17.0 234 of 413 Hy3 Preview (no reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 22.7 182 of 413 Hy3 Preview (thinking reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 73.2% 232 of 360 Hy3 Preview (no reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 86.7% 107 of 360 Hy3 Preview (thinking reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 7.0% 312 of 419 Hy3 Preview (no reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 27.8% 165 of 419 Hy3 Preview (thinking reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 48.0% 150 of 288 Hy3 Preview (no reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 63.1% 97 of 288 Hy3 Preview (thinking reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 42.0% 312 of 407 Hy3 Preview (no reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 64.7% 232 of 407 Hy3 Preview (thinking reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 31.8% 97 of 282 Hy3 Preview (no reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 34.1% 84 of 282 Hy3 Preview (thinking reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 67.5% 135 of 286 Hy3 Preview (no reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 92.7% 45 of 286 Hy3 Preview (thinking reasoning) 20 Oct 2026

Not shown: Not business work, and a blend: read the parts for any one task.

UGI Leaderboard · reported by UGI Leaderboard · 6 results
MeasureValueRankConfigurationDated
UGI: requested-length error% off the requested word count, lower is better 32.0% 272 of 379 Hy3 Preview (no reasoning) 2 May 2026
UGI: requested-length error% off the requested word count, lower is better 18.0% 158 of 379 Hy3 Preview (thinking reasoning) 2 May 2026
UGI: style adherencescore from 0 to 1, higher is better 0.35 210 of 379 Hy3 Preview (no reasoning) 2 May 2026
UGI: style adherencescore from 0 to 1, higher is better 0.33 268 of 379 Hy3 Preview (thinking reasoning) 2 May 2026
UGI: writing scorescore out of 100, higher is better 41.6 214 of 379 Hy3 Preview (no reasoning) 2 May 2026
UGI: writing scorescore out of 100, higher is better 51.1 161 of 379 Hy3 Preview (thinking reasoning) 2 May 2026

Not shown: Not other format limits such as character counts or bullet counts.

Test your own prompt.

Benchmarks aren't your task. Run your prompt on the models you could use, with every answer checked and costed.