Skip to content
ModelCap

Hy3 preview

All-versions archive #77Arena #123Restricted license

Tencent·tencent/hy3-preview

Hy3 preview is a language model published by Tencent. 1 OpenRouter endpoint is published in the latest snapshot. It accepts text input and returns text. The published context limit is 262K tokens. The catalogue declares support for reasoning, tool use.

ModelCap Index · All-versions archive
64.8#77
Measured · 65% support · score interval 64.8–74.4
Market Gravity
34.3
Secondary market signal

Superseded by Hy3. It no longer holds a current-board position; its all-versions archive rank is #77.

Input
$0.063
per 1M tokens
Output
$0.21
per 1M tokens
Context
262K
tokens
Providers
1
endpoint
Gateway spend
—
latest · Vercel share

Providers

Uptime measured over the last 30 minutes
ProviderInputOutputContextQuantUptime
GMICloudbf16$0.063$0.21262Kbf16100.0%

Reported evaluation leads (4)

Excluded from score

Structured rows reported in this exact model repository at the pinned revision below. They are useful research leads, but remain quarantined until dataset revision, harness, configuration and identity pass ModelCap’s independent admission review.

Dataset / taskMetricValueRevisionProvenance
cais/hle
hle
—30
Unpinned
not published
Reported
Quarantined
harborframework/terminal-bench-2.0
terminalbench_2
—54.4
Unpinned
not published
Reported
Quarantined
Idavidrein/gpqa
diamond
—87.2
Unpinned
not published
Reported
Quarantined
SWE-bench/SWE-bench_Verified
swe_bench_%_resolved
—74.4
Unpinned
not published
Reported
Quarantined

Other versions (1)

ModelCap Index · All-versions archive

Methodology
64.8
Measured · 2 public benchmark observations across 2 boards
All-versions archive
#77
Public rank basis
Measured (measured)
Index support
64.5% · strong
Index interval
64.8–74.4
Rank posterior
Not available
Top 5 / Top 10 probability
Not available / Not available
Posterior as of
Snapshot timestamp unavailable
Identity binding
Exact catalogue product · tencent/tencent/hy3-preview · aggregates disclosed benchmark configurations · endpoint configuration not claimed · artifact metadata tencent/Hy3-preview@549c2b3a0fd5b9a6c6059a9935bf0d59ab69d75a (not the evaluation revision)
Direct benchmark observations
2
Breadth-adjusted BES point estimate
69.6
Bayesian Evidence Score · measured evidence input only

BES normalizes admitted public benchmark evidence for measured Index rows. Its legacy score and observables prior do not define the public language rank.

Observed capability
69.6
Evidence status
confirmed · 78% mass
General preference
69.3
Coding
71.1
Agents & tools
—
Reasoning
—
Evidence breadth
92%
Evidence coverage
65%
Market and catalogue signals · secondary, never the language Index rank
Usage (OpenRouter popularity)
22.0
Liquidity (providers × uptime)
45.6
Open reach (HF downloads)
—
Surface (context / tools / modalities)
75.3
Freshness
66.5
  • · OpenRouter popularity #85
  • · 1 live provider
  • · Catalogue: context 262144, reasoning, tools
  • · First seen 2026-04-22

These adoption and deployment observations remain context only. They do not change this model's ModelCap Index score or rank.

BES evidence selected · measured input only

The bounded 0–100 capability score blends each source’s competitive placement with its published achievement, then combines capability families. Evidence breadth adds a modest uncertainty adjustment; price and popularity are not part of benchmark-led rank.

Deployment readiness

Metadata index
69.3
partial · 89% metadata coverage
Not capability
Artifact reproducibility
85.0
Access & legal clarity
67.5
Deployability
68.7
Evaluation provenance
48.3

Missing: artifactReproducibility.base-lineage, accessLegalClarity.resolvable-license-terms, evaluationProvenance.evaluation-dataset-revisions.

A metadata completeness and deployability index—not a safety certification, quality grade, or production-readiness claim.

Market activity

Venue coverage
Market Gravity
34.3
OpenRouter weekly popularity
#85 of 298

Market Gravity uses OpenRouter’s full-catalogue popularity order. Sparse Vercel top-ten observations appear only when published and do not affect the score.

Arena preference

Status
Ranked
Source
Arena
Official rank
#123 of 385
Rating
1411.9
95% confidence interval
1404–1419
Votes
7K
Source snapshot
6 Aug 2026, 13:00 UTC
Category
overall
Identity match confidence
98%

Market Gravity breakdown

Usage55%
22.0

OpenRouter weekly popularity across the full model catalogue.

Liquidity25%
45.6

Independent providers versus the model's open or closed cohort, weighted by uptime.

Open reach15%
50.0

Hugging Face 30-day downloads for open models; neutral for closed models.

Freshness5%
66.5

Time since first public availability, on a six-month half-life.

Specification

Context window
262K tokens
Max output
—
Inputs
Text
Outputs
Text
Tokenizer
Other
Cached input tokens
$0.021 / 1M
First seen on OpenRouter
22 Apr 2026

Capabilities

  • Supported: Reasoning
  • Supported: Tool use
  • Not supported: Structured output
  • Not supported: Response format
  • Not supported: Moderated

Weights & access

Restricted license. Downloadable weights use a model-specific or use-restricted license. Review the linked license before use, redistribution, or commercial deployment.

Access
Restricted license
Licence
other
Pinned revision
549c2b3a0fd5
Parameters
298.8B
Architecture
HYV3ForCausalLM
Model type
hy_v3
Hugging Face downloads (30d)
61K
Downloads (all time)
285K
Likes
301
Repository
tencent/Hy3-preview
Repository updated
23 Apr 2026, 14:59 UTC