TENCENT HY TEAM · FIELD VERIFIED

Tencent Hy4

Primary sources last checked on 21 Nov 2026. Verification freshness is tracked separately from factual change, and unknown values remain explicit rather than inferred.

NotableIndependent Artificial Analysis scores a named Tencent Hy4 service at 35, but checkpoint-parity with this exact downloadable SHA is not independently established; retain Notable until matched. First-party immutable source SHA, actual public weights and explicitly scoped license support editorial Notable relevance, not independent runtime performance.Curation method →

MODEL INTELLIGENCE · SOURCE-FIRST

What is Tencent Hy4?

OMW evidence review21 Nov 2026

Open Model Weights records Tencent Hy4 as a 770B language backbone / 49B active (publisher); 774.43B tensor values across released checkpoint-parameter text, image and video input; text output open-weight model in the Hy4 family, with Tencent Hy Team listed as the developer. The checked source repository provides 245 recognized weight artifacts, format evidence for Safetensors and a verified context window of 1,048,576 tokens, publisher-declared; no OMW runtime test. Values that are not evidenced in the checked sources remain explicitly undisclosed or unknown.

The repository declares Tencent Hy License v1.0. Custom Tencent Hy License terms and possible MaaS restrictions require use-case-specific review. Open Model Weights reports the checked license evidence and classification; this is not legal advice.

No base model is declared in the checked standard lineage metadata, so Open Model Weights does not infer one. Training recipe status: not-independently-reproduced; data disclosure status: unknown.

+ Save to my shortlistMonitor changes →Local browser monitoring · no background email alerts

03 · PRACTICAL TECHNICAL PROFILE

Deployment, licensing and reproducible evidence.

Inspect weight-memory estimates, model lineage and runtime evidence separately from promotional or untested performance claims.

DEPLOYMENT PROFILE

Weight memory, without pretending it is a deployment guarantee.

These bars visualize weight-only decimal-GB estimates. They exclude KV cache, activations, runtime overhead, optimizer state and sharding.

WEIGHT-ONLY MEMORY
BF16 / FP16~1,548.9 GB
FP8 / INT8~774.4 GB
INT4~387.2 GB
FIELD EVIDENCE MAP
WeightsVerified
LicenseVerified
Commercial useAllowed with conditions
ContextNot declared
FormatsVerified
LineageNot declared
Training recipeNot independently reproduced
Training dataUnknown
RuntimeSource-derived · not runtime-tested

LICENSE & USE

Tencent Hy License v1.0

Custom Tencent Hy License terms and possible MaaS restrictions require use-case-specific review

Declared licenseTencent Hy License v1.0
ClassificationAllowed with conditions
VerificationLicense file checked

This section summarizes the checked repository evidence. It is not legal advice. Inspect license evidence ↗

LINEAGE & PROVENANCE

What the checked sources disclose — and what they do not.

Open Model Weights does not infer a hidden base model or training corpus when the checked sources do not declare one.

LINEAGE
BASE / PARENTNot declared in model-card metadataUnknown
not inferred
CURRENT RECORDTencent Hy4Hy4
TRAINING DISCLOSURE
Training recipeNot independently reproduced
Training dataUnknown
FamilyHy4

RUNTIME SIGNALS

Compatibility claims stay attached to their evidence level.

Declared and model-card-mentioned runtimes are discovery signals. They are not treated as Open Model Weights runtime tests unless explicitly measured.

MentionedvLLM
MentionedSGLang
OMW runtime testNot performed

EVIDENCE TRAIL

Sources and observed history.

This is where the article remains auditable: source links on one side, the registry's observed change history on the other.

04 · EVIDENCE AND UPDATES

Verification, changes and claim checks.

Inspect dated source checks and any differences between published claims and the evidence; these checks are not independent model benchmarks.

Evidence statusSource-first verified
Last verified21 Nov 2026
Full field verification21 Nov 2026

SOURCE-BASED VISUAL BRIEF · NOTABLE

Tencent Hy4: technical evidence at a glance

Three evidence-linked visual summaries: parameter scale, illustrative precision memory and verifiable provenance. No performance ranking is implied.

01 / PARAMETER SCALE

Total versus active parameters

Total parameters774.4B
Active parameters · publisher49B
Repository-derived or publisher-described parameter count
Total and active parameters describe different published quantities, not inference speed or runtime memory.
02 / MEMORY ESTIMATE

Weight-only precision footprint

No reliable single-model weight-only memory figure.

An unavailable estimate or multi-component video system is not replaced with a guessed memory requirement.
03 / AUDIT TRAIL

From repository to record

  1. 01
    Source revisionPinned revision ↗
  2. 02
    Weight artifacts245 recognized artifacts · public access
  3. 03
    License declarationtencent-hy-license ↗License file checked
  4. 04
    Latest published source check
Repository SHA c10c0de5b6f8…; a source check is not an inference benchmark or legal clearance. See the per-model evidence JSON for the exact source record.

FRESHNESS & VERIFICATION

Current evidence, dated honestly.

A verification check does not pretend the model changed. Open Model Weights updates the verification date after a real primary-source check, while factual-update dates move only when retained source evidence changes.

Last verified
Latest primary-source check · source manifest manual review
Last factual update
No change observed
Baseline first observed 21 Nov 2026; no later factual change observed
Repository created
Source repository date; used only as a transparent release proxy when no stronger release date exists

CONFLICT INTELLIGENCE

Claims stay attached to evidence class.

No active comparable-source conflict detectedThis is not a claim that every possible source agrees; it covers the comparable fields currently checked.
Publisher claims5
Source observed4
OMW derived3
OMW measured0

05 · SOURCE RECORD

Exact model artifacts and technical fields.

Full repository-level attributes, licensing, format details and provenance are available below without hiding or rewriting the original data.

01 · WEIGHTS

Exact repository artifacts

Weightsverified-available
Accesspublic
Repositorytencent/Hy4
Recognized weight files245
Show all 245 exact weight files
HF API file listing ↗

02 · LICENSE

License & commercial use

Declared licenseTencent Hy License v1.0
License IDtencent-hy-license
Commercial useCustom Tencent Hy License terms and possible MaaS restrictions require use-case-specific review
Repository license filesLICENSE

Classification describes the checked license terms; it is not legal advice.

License / model-card evidence ↗

03 · HARDWARE

Weight-memory estimate

BF16 / FP16~1,548.9 GB
FP8 / INT8~774.4 GB
INT4~387.2 GB

Arithmetic for hypothetical whole-model uniform precision, not the exact published BF16 artifact. Official repository storage: 1548.94 GB including nonweight files; no runtime memory or speed measurements.

04 · FORMATS

Formats & precision evidence

Safetensors
BF16: observedFP16: not observedFP8: not observedINT8: not observedINT4: not observedGGUF: not observed

“Not observed” means no matching repository artifact/name/dtype signal was found; it does not prove no third-party conversion exists.

05 · LINEAGE

Base model & family

FamilyHy4
VariantFull instruct BF16
Declared base modelNot declared in model-card metadata
RelationNot declared

06 · TRAINING ASSETS

Recipe & data disclosure

Training recipenot-independently-reproduced
Data disclosureunknown
EvaluationsArtificial Analysis independently evaluates Tencent Hy4 as a named service at 35 points; parity with this exact pinned checkpoint and harness is not established. No OMW inference run.
Intermediate checkpointsNot independently reviewed

07 · RUNTIME SUPPORT

Declared, mentioned and externally evidenced runtimes

Declared librarytransformers
Declared/taggedNone
Mentioned in model cardvLLM, SGLang
Artifact signalsNone
External primary runtime evidenceNone
Tested by Open Model WeightsNo — source-derived support only

Open Model Weights separates model-repository evidence from external runtime-release evidence. An external support statement does not change the official model repository's weight formats and does not imply an Open Model Weights inference benchmark.

08 · SOURCES

Field evidence

Primary / declared sources6
Repository revisionc10c0de5b6f8
Source policyHow evidence is classified →

09 · VERIFICATION HISTORY

Checked, not assumed

First-party immutable source audit

Pinned c10c0de5b6f850703fb0a90e65d1c32b558b9167; recorded official 245 weight-file names and original source license; independent performance not inferred.

licenselicense-file-checked
commercial useclassified-per-source
contextpublisher-declared
weight filesverified-listed-original
formatsverified-safetensors
base modelnot-declared
training recipenot-independently-verified
data disclosurenot-independently-verified
runtime supportsource-described-not-tested

MODELS FOR AI AGENTS · ROLE EVIDENCE

Where this model fits in an agent system

Multimodal reasoning model. A candidate for a reasoning or coordination role. No end-to-end agent performance is inferred from its model category.

The curated notable tier is not an agent-readiness rating. Model-specific tool-call success, long-horizon task accuracy and compatible runtime configurations are not independently established here.

Source-linked license: Tencent Hy License v1.0 (allowed-with-conditions commercial-use classification; see original terms).

Context: 1048576 configured tokens.

BF16 weights: 1548.87 estimated decimal GB, not runtime VRAM.

Compare agent model candidates → · Agent evaluation method → · Inspect original model repository ↗

Agent-specific published evidence

These records are separately sourced. They do not replace independent model-plus-runtime tests.

No model-specific agent capability evaluation is recorded for this checkpoint. Missing evidence is unknown, not a failed task.

Read evidence details and benchmarks → · Use the agent model selector →