Open Model WeightsTENCENT HUNYUAN · FIELD VERIFIED
Primary sources last checked on 26 Oct 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 Hy-Nomas as a Multiple video components (no single verified total)-parameter text and camera trajectory to generated video open-weight model in the Hy-Nomas family, with Tencent Hunyuan listed as the developer. The checked source repository provides 27 recognized weight artifacts, format evidence for PyTorch, Safetensors and a context window that is not declared in the checked standard sources. Values that are not evidenced in the checked sources remain explicitly undisclosed or unknown.
The repository declares apache-2.0. Commercial use allowed under apache-2.0 conditions. 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-disclosed-in-checked-repository; data disclosure status: described-in-model-card.
+ Save to my shortlistMonitor changes →Local browser monitoring · no background email alerts
DIRECT ANSWER · FEATURED MODEL BRIEFING
Source-grounded editorial layer · reviewed 2026-10-26
Tencent Hy-Nomas is a first-party open-weight interactive video-world generation system released for segment-wise interactive video with text and camera conditioning. The publisher documents The source includes DiT generation weights, DCAE-V components, a PixelUMM decoder and a separate Qwen text encoder. This specific version must pass source-verified model-file, model-card, exact SHA and license checks before Open Model Weights publishes a field-verified profile. Its declared component-wise, no verified single model total and not a language-token context window are differentiated from measured tensors, runtime memory and independent performance. Vendor benchmark claims do not confer an automatic Leading designation. Open Model Weights independently checked the source revision, files and declared license for this canonical record; it has not executed model inference.
WHY IT MATTERS
The source includes DiT generation weights, DCAE-V components, a PixelUMM decoder and a separate Qwen text encoder.
The official architecture uses 54 dual-stream diffusion blocks and a persistent KV cache to extend generated video segments.
The publisher demonstrates textual scene edits and camera motion conditioning across successive generated video segments.
MODEL-SELECTION VIEW
The source includes DiT generation weights, DCAE-V components, a PixelUMM decoder and a separate Qwen text encoder. The source-of-truth is the original publisher checkpoint and its revision-pinned file manifest. Keep the technical observation separate from release marketing, hosted-service features, third-party comparisons and any later edits to the model card; this is a prerequisite for accurate open-weight AI discovery.
An aggregate artifact size is not a single logical parameter count; dependencies require independent revision and license inventories. A reproducible evaluation must retain the original runtime version, model and dependent component revisions, task prompts, negative examples, latency and actual peak memory. This page documents a practical verification protocol, not an Open Model Weights independent inference result or universal model ranking.
The official architecture uses 54 dual-stream diffusion blocks and a persistent KV cache to extend generated video segments. The source-of-truth is the original publisher checkpoint and its revision-pinned file manifest. Keep the technical observation separate from release marketing, hosted-service features, third-party comparisons and any later edits to the model card; this is a prerequisite for accurate open-weight AI discovery.
Preserved contextual frames and synchronized camera trajectories matter more here than a language model's maximum token count. A reproducible evaluation must retain the original runtime version, model and dependent component revisions, task prompts, negative examples, latency and actual peak memory. This page documents a practical verification protocol, not an Open Model Weights independent inference result or universal model ranking.
The publisher demonstrates textual scene edits and camera motion conditioning across successive generated video segments. The source-of-truth is the original publisher checkpoint and its revision-pinned file manifest. Keep the technical observation separate from release marketing, hosted-service features, third-party comparisons and any later edits to the model card; this is a prerequisite for accurate open-weight AI discovery.
Evaluate object consistency, unwanted transformations and movement continuity with captured event timing and trajectory matrices. A reproducible evaluation must retain the original runtime version, model and dependent component revisions, task prompts, negative examples, latency and actual peak memory. This page documents a practical verification protocol, not an Open Model Weights independent inference result or universal model ranking.
The 8-step distilled setup targets 720×1280 video rendered at 24 output frames per second under published inference parameters. The source-of-truth is the original publisher checkpoint and its revision-pinned file manifest. Keep the technical observation separate from release marketing, hosted-service features, third-party comparisons and any later edits to the model card; this is a prerequisite for accurate open-weight AI discovery.
Playback FPS must not be misrepresented as actual wall-clock real-time generation speed on commodity devices. A reproducible evaluation must retain the original runtime version, model and dependent component revisions, task prompts, negative examples, latency and actual peak memory. This page documents a practical verification protocol, not an Open Model Weights independent inference result or universal model ranking.
Standard inference uses eight-way sequence parallelism, while a special single-4090 mode combines INT8 DiT, FP8 decoder and CPU offload. The source-of-truth is the original publisher checkpoint and its revision-pinned file manifest. Keep the technical observation separate from release marketing, hosted-service features, third-party comparisons and any later edits to the model card; this is a prerequisite for accurate open-weight AI discovery.
Those execution paths are not interchangeable with full precision; benchmark memory and wall time with exactly the documented component versions. A reproducible evaluation must retain the original runtime version, model and dependent component revisions, task prompts, negative examples, latency and actual peak memory. This page documents a practical verification protocol, not an Open Model Weights independent inference result or universal model ranking.
The repository declares Apache-2.0 for publisher work and supplies Hy-Nomas-Bench with 280 interactive cases. The source-of-truth is the original publisher checkpoint and its revision-pinned file manifest. Keep the technical observation separate from release marketing, hosted-service features, third-party comparisons and any later edits to the model card; this is a prerequisite for accurate open-weight AI discovery.
Third-party encoders retain their own terms, and first-party evaluation tasks are not independent proof of physically valid simulation. A reproducible evaluation must retain the original runtime version, model and dependent component revisions, task prompts, negative examples, latency and actual peak memory. This page documents a practical verification protocol, not an Open Model Weights independent inference result or universal model ranking.
IMPLEMENTATION & EVALUATION
Four-step original-source test plan for Tencent Hy-Nomas, not an independently executed performance benchmark.
Record exact source SHA, original artifacts and Apache-2.0 (publisher-owned assets) license text with upstream notices.
Immutable manifest with official revision and compatible dependency list
Upstream repository can change after intake
Run Tencent Hy-Nomas through its original official task interface, retaining model and dependency versions.
Repeatable source-aligned output type and stable input handling
Generic family examples may not match this checkpoint
Score task-specific positive and negative cases for segment-wise interactive video with text and camera conditioning, with published failure examples.
Held-out task quality table with traceable observed errors
Publisher benchmark numbers are not independent results
Record throughput, p95 latency, peak memory, operating system and rollback criteria under fixed tasks.
Written deployment decision with real observed resource metrics
Weight-only arithmetic omits cache, activations and extra decoder assets
Proposed tests, not OMW benchmark results. Read the official model card at the pinned source revision ↗ Inspect OMW field evidence ↗
PRIMARY & INDEPENDENT SOURCES
FAQ
Tencent Hy-Nomas is a publisher-released interactive video-world generation system intended for segment-wise interactive video with text and camera conditioning. Its official documentation describes The source includes DiT generation weights, DCAE-V components, a PixelUMM decoder and a separate Qwen text encoder. Open Model Weights verifies source records rather than inventing comparative scores.
The official publisher maintains a Hugging Face checkpoint with a model card and weight-file inventory. Its exact source SHA and artifact accessibility must pass a targeted verifier before a canonical record can be called field-verified.
The official repository declares Apache-2.0 (publisher-owned assets), but the exact revision's license file, dependency scope and downstream obligations must be checked. The public-weight label does not grant blanket permission for any conceivable deployment.
Notable reflects source-released specialist significance, not an absolute AI leaderboard result. Independent checkpoint-equivalent benchmark results are not currently sufficient to award Leading status for this record.
There is no universally verified device-memory minimum for this deployment. Stored parameters, precision, caches, external decoders, tensor parallelism and inference frameworks all affect the actual peak resident footprint.
Pin a 40-character source SHA and matching model and component files, reproduce the official task interface and test against held-out positive and negative cases. Record real device memory, latency, quality failures and license conditions.
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
Commercial use allowed under apache-2.0 conditions
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.
Checked Hugging Face API metadata, exact repository file list, model card and config when accessible.
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 comparable single-model parameter total is available.
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
DiT/base_ar/diffusion_pytorch_model_fp32.safetensorsDiT/distilled/diffusion_pytorch_model_ema_0_995.safetensorsDiT/distilled_int8/diffusion_pytorch_model_ema_0_995_int8.safetensorsDiT/lora_weights/cinematic/adapter.safetensorsDiT/lora_weights/dual_arm_robotics/adapter.safetensorsPE/model-00001-of-00018.safetensorsPE/model-00002-of-00018.safetensorsPE/model-00003-of-00018.safetensorsPE/model-00004-of-00018.safetensorsPE/model-00005-of-00018.safetensorsPE/model-00006-of-00018.safetensorsPE/model-00007-of-00018.safetensorsPE/model-00008-of-00018.safetensorsPE/model-00009-of-00018.safetensorsPE/model-00010-of-00018.safetensorsPE/model-00011-of-00018.safetensorsPE/model-00012-of-00018.safetensorsPE/model-00013-of-00018.safetensorsPE/model-00014-of-00018.safetensorsPE/model-00015-of-00018.safetensorsPE/model-00016-of-00018.safetensorsPE/model-00017-of-00018.safetensorsPE/model-00018-of-00018.safetensorsPE/model.safetensors.index.jsonVAE/DCAE-V/dc-ae-v-1.0-f32t4c32-bf16.ptVAE/PixelUMM_FT/diffusion_pytorch_model.safetensorsVAE/PixelUMM_FT_FP8/diffusion_pytorch_model_fp8.safetensors02 · LICENSE
Classification describes the checked license terms; it is not legal advice.
License / model-card evidence ↗03 · HARDWARE
Publisher uses multiple DiT, encoder and VAE components. Standard 8-way sequence parallelism differs from specific single-GPU INT8/FP8 offload.
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
Checked Hugging Face API metadata, exact repository file list, model card and config when accessible.
MODELS FOR AI AGENTS · ROLE EVIDENCE
Interactive video world generator. A potential media-processing tool invoked by a separate controller; image/video/speech generation is not agent autonomy.
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: apache-2.0 (allowed commercial-use classification; see original terms).
Context: Unknown or not declared.
BF16 weights: Not calculated.
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 →