← Model list

Hy3 preview

tencent·tencent/hy3-preview·GA·Open weights·Hy series

Tencent Hy reasoning model for coding, instruction following, and agent tasks

At a glance
Official price$0 / $0 per 1M
Cache read $0 · Tencent TokenHub
Lowest paid$0.066 / $0.26
SiliconFlow Cloud · 2.7× spread
1 more $0 channels
Context256,000
⚠ Providers report 256,000–262,144; the table below is authoritative
Output limit64,000
Capabilities
ReasoningTool useStructured outputTemperatureAttachments
⚠ Providers report capability flags inconsistently
Modalities
Text
Knowledge cutoff
Released / updated2026-04-20 / 2026-04-20

Quality & performanceArtificial Analysis · Intelligence Index v4.1 · rep. tier Reasoning

Intelligence34.4
Value300
Coding
Agentic
Output speed
TTFT
Value formulaIQ 34.4 ÷ min blended $0.115 = 300
Reasoning tier → intelligence / speed (higher tier = stronger but slower)
Non-reasoningIQ 26.6
ReasoningIQ 34.4

Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.

Available at 6 providers5 with public prices · 1 subscription-covered

ProviderTierInputOutputCache readCache writeContextOutput limitStatus
Tencent TokenHubOfficialFirst-partySubscription$0$0256,00064,000
SiliconFlow
tencent/Hy3-preview
Cloud$0.066$0.26$0.029262,144262,144
AIHubMixGateway$0.17$0.567$0.051256,000128,000
ZenMuxGateway$0.172$0.572$0.058$0256,00064,000
OpenRouterGateway$0.18$0.60$0.06262,144262,144
Kilo GatewayGateway$0.18$0.60$0.06262,144262,144

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Your usage cost

1SiliconFlow$21.76
2AIHubMix$48.05
3ZenMux$49.32
4OpenRouter$51.60
5Kilo Gateway$51.60

1 more channels offer $0 (Tencent TokenHub); free tiers usually have rate limits and no SLA, excluded from ranking.

The cheapest paid channel is the only channel.

Benchmark1 items

NameConditionsScoreMetricSource
SWE-Bench Verified74.4resolvedSource ↗

The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.

Reasoning control

Toggle (on / off)effort = loweffort = mediumeffort = highbudget_tokens ≥ 128 ≤ 32,768effort = none

Related models

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$0.115$02026-08-052026-08-162026-08-05 · Input list $02026-08-06 · Input list $02026-08-07 · Input list $02026-08-08 · Input list $02026-08-09 · Input list $02026-08-10 · Input list $02026-08-11 · Input list $02026-08-12 · Input list $02026-08-13 · Input list $02026-08-16 · Input list $02026-08-05 · Output list $02026-08-06 · Output list $02026-08-07 · Output list $02026-08-08 · Output list $02026-08-09 · Output list $02026-08-10 · Output list $02026-08-11 · Output list $02026-08-12 · Output list $02026-08-13 · Output list $02026-08-16 · Output list $02026-08-05 · Min blended $0.102026-08-06 · Min blended $0.102026-08-07 · Min blended $0.102026-08-08 · Min blended $0.102026-08-09 · Min blended $0.102026-08-10 · Min blended $0.102026-08-11 · Min blended $0.102026-08-12 · Min blended $0.102026-08-13 · Min blended $0.102026-08-16 · Min blended $0.115