← Model list

Hy3

tencent·tencent/hy3-2·GA·Open weights·Hy series·NEW

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$0.126 / $0.522
Requesty Gateway · 1.3× spread
2 more $0 channels
Context256,000
⚠ Providers report 202,752–262,144; the table below is authoritative
Output limit64,000
Capabilities
ReasoningTool useStructured outputTemperatureAttachments
Modalities
Text
Knowledge cutoff
Released / updated2026-07-06 / 2026-07-06

Quality & performanceArtificial Analysis · Intelligence Index v4.1

Intelligence42.2
Value188
Coding58.8
Agentic31.4
Output speed67 tok/s
TTFT2.88 s
Cost per task$0.0357
Value formulaIQ 42.2 ÷ min blended $0.225 = 188

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

Available at 11 providers9 with public prices · 2 subscription-covered

ProviderTierInputOutputCache readCache writeContextOutput limitStatus
Tencent TokenHubOfficial
hy3
First-partySubscription$0$0256,00064,000
Tencent Token Plan
hy3
GatewaySubscription$0$0256,00064,000
Requesty
hy3
Gateway$0.126$0.522$0.032262,144262,144
OpenRouter
tencent/hy3
Gateway$0.132$0.528$0.033262,144128,000
Vercel AI Gateway
tencent/hy3
Cloud$0.132$0.528$0.033256,000128,000
Jalapeno Cloud
Hy3
Gateway$0.14$0.58202,75264,000
Deep Infra
tencent/Hy3
Cloud$0.14$0.58$0.035262,14464,000
Hugging Face
tencent/Hy3
Cloud$0.14$0.58262,14464,000
DevPass (LLM Gateway)
hy3
Gateway$0.14$0.58$0.035262,14464,000
Kilo Gateway
tencent/hy3
Gateway$0.14$0.58$0.035262,144128,000
CrossModel
tencent/hy3
Gateway$0.16$0.64$0.04$0.16262,144262,144

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

1Requesty$39.96
2OpenRouter$40.92
3Vercel AI Gateway$40.92
4Deep Infra$44.40
5DevPass (LLM Gateway)$44.40
6Kilo Gateway$44.40

2 more channels offer $0 (Tencent TokenHub, Tencent Token Plan); 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 Verified78resolvedSource ↗

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 = higheffort = noneeffort = maxbudget_tokens

1 / 11 providers expose no reasoning control (reasoning_options: []).

Related models

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$0.26$02026-08-052026-08-212026-08-05 · Input list $0.0662026-08-06 · Input list $0.0662026-08-07 · Input list $0.0662026-08-08 · Input list $0.0662026-08-09 · Input list $0.0662026-08-10 · Input list $0.0662026-08-11 · Input list $0.0662026-08-12 · Input list $0.0662026-08-13 · Input list $02026-08-21 · Input list $02026-08-05 · Output list $0.262026-08-06 · Output list $0.262026-08-07 · Output list $0.262026-08-08 · Output list $0.262026-08-09 · Output list $0.262026-08-10 · Output list $0.262026-08-11 · Output list $0.262026-08-12 · Output list $0.262026-08-13 · Output list $02026-08-21 · Output list $02026-08-05 · Min blended $0.1152026-08-06 · Min blended $0.1152026-08-07 · Min blended $0.1152026-08-08 · Min blended $0.1152026-08-09 · Min blended $0.1152026-08-10 · Min blended $0.1152026-08-11 · Min blended $0.1152026-08-12 · Min blended $0.1152026-08-13 · Min blended $0.2312026-08-21 · Min blended $0.225