Pin official model source
Upstream repositories and LICENSE files can change.
Record full 40-character SHA, visible original checkpoint names and pinned card and config URLs.
Silent source drift or mixing different variants.
TENCENT HY TEAM · FIELD VERIFIED
Primary sources last checked on 21 Nov 2026. Verification freshness is tracked separately from factual change, and unknown values remain explicit rather than inferred.
MODEL INTELLIGENCE · SOURCE-FIRST
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
DIRECT ANSWER · FEATURED MODEL BRIEFING
Source-grounded editorial layer · reviewed 2026-11-21
Tencent Hy4 is an original, publicly downloadable 770B language backbone / 49B active (publisher); 774.43B tensor values across released checkpoint release by Tencent Hy Team. Its official Hugging Face repository is pinned at c10c0de5b6f850703fb0a90e65d1c32b558b9167. The declared license is Tencent Hy License v1.0 with a standalone LICENSE file. Open Model Weights has not executed inference or benchmarks for this checkpoint.
WHY IT MATTERS
Publisher describes a sparse language MoE with an associated vision component and separate speculative drafter; parameter scopes must not be mixed.
The exact source-reported checkpoint contains 245 Safetensors files at the pinned revision, with 774.43 billion observed tensor values. This is a storage metadata count, not a runtime-memory or capability benchmark.
Custom Tencent Hy v1.0 license has additional terms, including large MaaS conditions. Independently measured Tencent Hy4 service scores are not shown to match this pinned downloadable artifact.
MODEL-SELECTION VIEW
Tencent Hy4 is an original, publicly downloadable 770B language backbone / 49B active (publisher); 774.43B tensor values across released checkpoint release by Tencent Hy Team. Its official Hugging Face repository is pinned at c10c0de5b6f850703fb0a90e65d1c32b558b9167. The declared license is Tencent Hy License v1.0 with a standalone LICENSE file. Open Model Weights has not executed inference or benchmarks for this checkpoint.
Publisher describes a sparse language MoE with an associated vision component and separate speculative drafter; parameter scopes must not be mixed.
Original checkpoint evidence is tied to https://huggingface.co/tencent/Hy4/tree/c10c0de5b6f850703fb0a90e65d1c32b558b9167. The publisher's source inventory lists 245 Safetensors files, including model-00001-of-00245.safetensors and model-00245-of-00245.safetensors. This source review inventories file names and aggregate tensor metadata, not a full local checksum download or inference run.
Parameter labels and published precision must be separated from activation counts and load-time memory. 770B language backbone / 49B active (publisher); 774.43B tensor values across released checkpoint.
The precise license classification is Tencent Hy License v1.0. A standalone LICENSE is present at the official pinned revision. Weight rights do not automatically license every dependency or dataset.
No independently verified base-model lineage was established by this source review.
Artificial Analysis independently evaluates Tencent Hy4 as a named service at 35 points; parity with this exact pinned checkpoint and harness is not established.
Custom Tencent Hy v1.0 license has additional terms, including large MaaS conditions. Independently measured Tencent Hy4 service scores are not shown to match this pinned downloadable artifact.
Candidate workflows: Multimodal long-context instruction following, enterprise workflows and agent research. Published runtime references mention vLLM, SGLang. Running vendor-specific loaders may execute source code; inspect and pin dependencies.
Repeatable evaluation should pin the exact revision, model processor and chat/decision template; publish hardware, memory, latency, task outcomes and failure cases. No such Open Model Weights runtime measurements are claimed here.
IMPLEMENTATION & EVALUATION
Source-pinned field evidence; no OMW model inference or independently measured throughput.
Upstream repositories and LICENSE files can change.
Record full 40-character SHA, visible original checkpoint names and pinned card and config URLs.
Silent source drift or mixing different variants.
Downloadable weights do not automatically cover runtime code and commercial rights.
Inspect source-pinned license and any publisher usage policy, third-party notices or remote model code.
Licensing overreach or arbitrary code execution.
Storage and publisher benchmark numbers are not production capabilities.
Publish exact hardware, precision, runtime, batch settings, memory and latency for a held-out task.
Overclaimed context, memory or performance.
Leading requires an independently evaluated checkpoint-identical reference.
Link independent result to the actual source checkpoint, harness and evaluation date.
Unjustified Leading promotion based on API-only or publisher scores.
Proposed tests, not OMW benchmark results. Read the official model card at the pinned source revision ↗ Inspect OMW field evidence ↗
PRIMARY & INDEPENDENT SOURCES
FAQ
The official tencent/Hy4 repository declares public, ungated Safetensors weights at immutable SHA c10c0de5b6f850703fb0a90e65d1c32b558b9167; the checked manifest lists 245 original files.
No. This is an evidence-based source audit, not an OMW execution or independent performance ranking.
The recorded publisher license is Tencent Hy License v1.0. A first-party LICENSE file was listed. Check all conditions and separately licensed components.
The repository storage is approximately 1548.94 decimal GB, not a measured GPU memory budget. Runtime operators, KV cache and media preprocessing add requirements.
Notable records differentiated primary-source releases; Leading requires matched, independent evidence for this exact checkpoint.
03 · PRACTICAL TECHNICAL PROFILE
Inspect weight-memory estimates, model lineage and runtime evidence separately from promotional or untested performance claims.
DEPLOYMENT PROFILE
These bars visualize weight-only decimal-GB estimates. They exclude KV cache, activations, runtime overhead, optimizer state and sharding.
LICENSE & USE
Custom Tencent Hy License terms and possible MaaS restrictions require use-case-specific review
This section summarizes the checked repository evidence. It is not legal advice. Inspect license evidence ↗
LINEAGE & PROVENANCE
Open Model Weights does not infer a hidden base model or training corpus when the checked sources do not declare one.
RUNTIME SIGNALS
Declared and model-card-mentioned runtimes are discovery signals. They are not treated as Open Model Weights runtime tests unless explicitly measured.
EVIDENCE TRAIL
This is where the article remains auditable: source links on one side, the registry's observed change history on the other.
Pinned c10c0de5b6f850703fb0a90e65d1c32b558b9167; recorded official 245 weight-file names and original source license; independent performance not inferred.
04 · EVIDENCE AND UPDATES
Inspect dated source checks and any differences between published claims and the evidence; these checks are not independent model benchmarks.
SOURCE-BASED VISUAL BRIEF · NOTABLE
Three evidence-linked visual summaries: parameter scale, illustrative precision memory and verifiable provenance. No performance ranking is implied.
No reliable single-model weight-only memory figure.
FRESHNESS & VERIFICATION
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.
CONFLICT INTELLIGENCE
05 · SOURCE RECORD
Full repository-level attributes, licensing, format details and provenance are available below without hiding or rewriting the original data.
01 · WEIGHTS
model-00001-of-00245.safetensorsmodel-00002-of-00245.safetensorsmodel-00003-of-00245.safetensorsmodel-00004-of-00245.safetensorsmodel-00005-of-00245.safetensorsmodel-00006-of-00245.safetensorsmodel-00007-of-00245.safetensorsmodel-00008-of-00245.safetensorsmodel-00009-of-00245.safetensorsmodel-00010-of-00245.safetensorsmodel-00011-of-00245.safetensorsmodel-00012-of-00245.safetensorsmodel-00013-of-00245.safetensorsmodel-00014-of-00245.safetensorsmodel-00015-of-00245.safetensorsmodel-00016-of-00245.safetensorsmodel-00017-of-00245.safetensorsmodel-00018-of-00245.safetensorsmodel-00019-of-00245.safetensorsmodel-00020-of-00245.safetensorsmodel-00021-of-00245.safetensorsmodel-00022-of-00245.safetensorsmodel-00023-of-00245.safetensorsmodel-00024-of-00245.safetensorsmodel-00025-of-00245.safetensorsmodel-00026-of-00245.safetensorsmodel-00027-of-00245.safetensorsmodel-00028-of-00245.safetensorsmodel-00029-of-00245.safetensorsmodel-00030-of-00245.safetensorsmodel-00031-of-00245.safetensorsmodel-00032-of-00245.safetensorsmodel-00033-of-00245.safetensorsmodel-00034-of-00245.safetensorsmodel-00035-of-00245.safetensorsmodel-00036-of-00245.safetensorsmodel-00037-of-00245.safetensorsmodel-00038-of-00245.safetensorsmodel-00039-of-00245.safetensorsmodel-00040-of-00245.safetensorsmodel-00041-of-00245.safetensorsmodel-00042-of-00245.safetensorsmodel-00043-of-00245.safetensorsmodel-00044-of-00245.safetensorsmodel-00045-of-00245.safetensorsmodel-00046-of-00245.safetensorsmodel-00047-of-00245.safetensorsmodel-00048-of-00245.safetensorsmodel-00049-of-00245.safetensorsmodel-00050-of-00245.safetensorsmodel-00051-of-00245.safetensorsmodel-00052-of-00245.safetensorsmodel-00053-of-00245.safetensorsmodel-00054-of-00245.safetensorsmodel-00055-of-00245.safetensorsmodel-00056-of-00245.safetensorsmodel-00057-of-00245.safetensorsmodel-00058-of-00245.safetensorsmodel-00059-of-00245.safetensorsmodel-00060-of-00245.safetensorsmodel-00061-of-00245.safetensorsmodel-00062-of-00245.safetensorsmodel-00063-of-00245.safetensorsmodel-00064-of-00245.safetensorsmodel-00065-of-00245.safetensorsmodel-00066-of-00245.safetensorsmodel-00067-of-00245.safetensorsmodel-00068-of-00245.safetensorsmodel-00069-of-00245.safetensorsmodel-00070-of-00245.safetensorsmodel-00071-of-00245.safetensorsmodel-00072-of-00245.safetensorsmodel-00073-of-00245.safetensorsmodel-00074-of-00245.safetensorsmodel-00075-of-00245.safetensorsmodel-00076-of-00245.safetensorsmodel-00077-of-00245.safetensorsmodel-00078-of-00245.safetensorsmodel-00079-of-00245.safetensorsmodel-00080-of-00245.safetensorsmodel-00081-of-00245.safetensorsmodel-00082-of-00245.safetensorsmodel-00083-of-00245.safetensorsmodel-00084-of-00245.safetensorsmodel-00085-of-00245.safetensorsmodel-00086-of-00245.safetensorsmodel-00087-of-00245.safetensorsmodel-00088-of-00245.safetensorsmodel-00089-of-00245.safetensorsmodel-00090-of-00245.safetensorsmodel-00091-of-00245.safetensorsmodel-00092-of-00245.safetensorsmodel-00093-of-00245.safetensorsmodel-00094-of-00245.safetensorsmodel-00095-of-00245.safetensorsmodel-00096-of-00245.safetensorsmodel-00097-of-00245.safetensorsmodel-00098-of-00245.safetensorsmodel-00099-of-00245.safetensorsmodel-00100-of-00245.safetensorsmodel-00101-of-00245.safetensorsmodel-00102-of-00245.safetensorsmodel-00103-of-00245.safetensorsmodel-00104-of-00245.safetensorsmodel-00105-of-00245.safetensorsmodel-00106-of-00245.safetensorsmodel-00107-of-00245.safetensorsmodel-00108-of-00245.safetensorsmodel-00109-of-00245.safetensorsmodel-00110-of-00245.safetensorsmodel-00111-of-00245.safetensorsmodel-00112-of-00245.safetensorsmodel-00113-of-00245.safetensorsmodel-00114-of-00245.safetensorsmodel-00115-of-00245.safetensorsmodel-00116-of-00245.safetensorsmodel-00117-of-00245.safetensorsmodel-00118-of-00245.safetensorsmodel-00119-of-00245.safetensorsmodel-00120-of-00245.safetensorsmodel-00121-of-00245.safetensorsmodel-00122-of-00245.safetensorsmodel-00123-of-00245.safetensorsmodel-00124-of-00245.safetensorsmodel-00125-of-00245.safetensorsmodel-00126-of-00245.safetensorsmodel-00127-of-00245.safetensorsmodel-00128-of-00245.safetensorsmodel-00129-of-00245.safetensorsmodel-00130-of-00245.safetensorsmodel-00131-of-00245.safetensorsmodel-00132-of-00245.safetensorsmodel-00133-of-00245.safetensorsmodel-00134-of-00245.safetensorsmodel-00135-of-00245.safetensorsmodel-00136-of-00245.safetensorsmodel-00137-of-00245.safetensorsmodel-00138-of-00245.safetensorsmodel-00139-of-00245.safetensorsmodel-00140-of-00245.safetensorsmodel-00141-of-00245.safetensorsmodel-00142-of-00245.safetensorsmodel-00143-of-00245.safetensorsmodel-00144-of-00245.safetensorsmodel-00145-of-00245.safetensorsmodel-00146-of-00245.safetensorsmodel-00147-of-00245.safetensorsmodel-00148-of-00245.safetensorsmodel-00149-of-00245.safetensorsmodel-00150-of-00245.safetensorsmodel-00151-of-00245.safetensorsmodel-00152-of-00245.safetensorsmodel-00153-of-00245.safetensorsmodel-00154-of-00245.safetensorsmodel-00155-of-00245.safetensorsmodel-00156-of-00245.safetensorsmodel-00157-of-00245.safetensorsmodel-00158-of-00245.safetensorsmodel-00159-of-00245.safetensorsmodel-00160-of-00245.safetensorsmodel-00161-of-00245.safetensorsmodel-00162-of-00245.safetensorsmodel-00163-of-00245.safetensorsmodel-00164-of-00245.safetensorsmodel-00165-of-00245.safetensorsmodel-00166-of-00245.safetensorsmodel-00167-of-00245.safetensorsmodel-00168-of-00245.safetensorsmodel-00169-of-00245.safetensorsmodel-00170-of-00245.safetensorsmodel-00171-of-00245.safetensorsmodel-00172-of-00245.safetensorsmodel-00173-of-00245.safetensorsmodel-00174-of-00245.safetensorsmodel-00175-of-00245.safetensorsmodel-00176-of-00245.safetensorsmodel-00177-of-00245.safetensorsmodel-00178-of-00245.safetensorsmodel-00179-of-00245.safetensorsmodel-00180-of-00245.safetensorsmodel-00181-of-00245.safetensorsmodel-00182-of-00245.safetensorsmodel-00183-of-00245.safetensorsmodel-00184-of-00245.safetensorsmodel-00185-of-00245.safetensorsmodel-00186-of-00245.safetensorsmodel-00187-of-00245.safetensorsmodel-00188-of-00245.safetensorsmodel-00189-of-00245.safetensorsmodel-00190-of-00245.safetensorsmodel-00191-of-00245.safetensorsmodel-00192-of-00245.safetensorsmodel-00193-of-00245.safetensorsmodel-00194-of-00245.safetensorsmodel-00195-of-00245.safetensorsmodel-00196-of-00245.safetensorsmodel-00197-of-00245.safetensorsmodel-00198-of-00245.safetensorsmodel-00199-of-00245.safetensorsmodel-00200-of-00245.safetensorsmodel-00201-of-00245.safetensorsmodel-00202-of-00245.safetensorsmodel-00203-of-00245.safetensorsmodel-00204-of-00245.safetensorsmodel-00205-of-00245.safetensorsmodel-00206-of-00245.safetensorsmodel-00207-of-00245.safetensorsmodel-00208-of-00245.safetensorsmodel-00209-of-00245.safetensorsmodel-00210-of-00245.safetensorsmodel-00211-of-00245.safetensorsmodel-00212-of-00245.safetensorsmodel-00213-of-00245.safetensorsmodel-00214-of-00245.safetensorsmodel-00215-of-00245.safetensorsmodel-00216-of-00245.safetensorsmodel-00217-of-00245.safetensorsmodel-00218-of-00245.safetensorsmodel-00219-of-00245.safetensorsmodel-00220-of-00245.safetensorsmodel-00221-of-00245.safetensorsmodel-00222-of-00245.safetensorsmodel-00223-of-00245.safetensorsmodel-00224-of-00245.safetensorsmodel-00225-of-00245.safetensorsmodel-00226-of-00245.safetensorsmodel-00227-of-00245.safetensorsmodel-00228-of-00245.safetensorsmodel-00229-of-00245.safetensorsmodel-00230-of-00245.safetensorsmodel-00231-of-00245.safetensorsmodel-00232-of-00245.safetensorsmodel-00233-of-00245.safetensorsmodel-00234-of-00245.safetensorsmodel-00235-of-00245.safetensorsmodel-00236-of-00245.safetensorsmodel-00237-of-00245.safetensorsmodel-00238-of-00245.safetensorsmodel-00239-of-00245.safetensorsmodel-00240-of-00245.safetensorsmodel-00241-of-00245.safetensorsmodel-00242-of-00245.safetensorsmodel-00243-of-00245.safetensorsmodel-00244-of-00245.safetensorsmodel-00245-of-00245.safetensors02 · LICENSE
Classification describes the checked license terms; it is not legal advice.
License / model-card evidence ↗03 · HARDWARE
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
“Not observed” means no matching repository artifact/name/dtype signal was found; it does not prove no third-party conversion exists.
05 · LINEAGE
06 · TRAINING ASSETS
07 · RUNTIME SUPPORT
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
09 · VERIFICATION HISTORY
Pinned c10c0de5b6f850703fb0a90e65d1c32b558b9167; recorded official 245 weight-file names and original source license; independent performance not inferred.
MODELS FOR AI AGENTS · ROLE EVIDENCE
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 ↗
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 →