Skip to main content

Integrate next

Living candidate queue for new adapters. This is not a commitment and it is not the v0.1.0 product roadmap.

FileRole
This pageCanonical list (git + docs site)
ROADMAP.mdProduct slices (train / eval / one local VTON / agents)
.cursor/skills/integrate-model/How to integrate once you pick a row

Surveyed: 29 August 2026 · Sources: build.nvidia.com/models, Google virtual-try-on-001, Alibaba OutfitAnyone-Plus, Photoroom Virtual Try-On, CatVTON, Leffa, Nemotron, Cosmos 3.

How to use: pick a next row → follow the integrate-model skill (Path A first-party API, Path B local). After ship, move the row to Shipped and bump the date.

StatusMeaning
nextWorth integrating when someone is ready
watchInteresting; wait for a clearer API or fashion fit
blockedNo first-party API / weights / license yet
skipOut of scope or we already cover it another way

NVIDIA access is usually NVIDIA_API_KEY on build.nvidia.com (hosted NIM) and/or a self-hosted NIM container. Prefer the hosted NIM for Path A; HF weights + Diffusers/vLLM for Path B. Do not wrap FLUX / Qwen-Image / Ideogram a second time through NIM — we already have first-party adapters.

OpenTryOn services today: vton · generate · edit · understand · video-generate · bg-remove. Rows marked new service need a new CLI/MCP category before an adapter.


Wave 1 — NVIDIA (fits existing services)​

Highest leverage: one NIM provider key unlocks understand + video. Nemotron is not a T2I/VTON family — it is open multimodal understanding / agents. Cosmos is NVIDIA’s generation stack.

CandidateCategoryPathSuggested idWhyStatus
Nemotron 3 Nano Omni 30B-A3B ReasoningUnderstand (image + video + audio + text)A NIMnemotron-omniHosted NIM chat. Path B local weights later.shipped
Nemotron 3.5 Super VL 120B-A12BUnderstand (image + video + text, agentic reasoning)A NIMnemotron-3.5-super-vlGA Oct 2026 on build.nvidia.com; no audio (use Omni).shipped
Cosmos 3 Generator (nano, 8B)Text-to-video, image-to-videoA NIM (POST infer)cosmos3T2V if prompt only; I2V if image set. Optional COSMOS3_INFER_URL.shipped
Cosmos 3 Reasoner (nano)Understand (physical / video)A NIMcosmos3-reasonerWorld-model VLM.shipped
Cosmos 3 Nano local weightsT2V / I2VB Diffuserscosmos3-localnvidia/Cosmos3-Nano, Cosmos3OmniPipeline, BF16/CUDA. Shipped Oct 2026.shipped
Nemotron Nano 12B V2 VLUnderstand (image + video)A NIM—Predecessor to Omni. Only if Omni is too large.watch
Cosmos 3 Generator super (32B)T2V / I2VA NIM (NIM_MODEL_SIZE=super)cosmos3-superSame API as nano; heavier GPU.watch
Cosmos Predict 2.5 2BT2V / I2V / V2VA NIMcosmos-predict-2.5Older WFM; prefer Cosmos 3 unless Predict-only features are needed.watch
Nemotron 3 Nano / Super / Ultra (text)Text-only agentsA NIM—Coding/planning MoE. No image/video I/O. Planner already uses other LLMs.skip
Nemotron OCR / Parse / page-tableDocument OCRA NIM—Useful later for PDP/catalog agents, not invoke-layer media.watch
Nemotron 3.5 Content SafetySafety classifierA NIM—Guardrail, not a generation/understand tool.watch
Nemotron 3.5 Super VL 120B-A12BUnderstand (image + video + text, agentic reasoning)A NIMnemotron-3.5-super-vlHybrid Mamba–Transformer MoE; 1M context; GA on build.nvidia.com. HF weights for self-host.shipped

Nemotron 3 family (context, Aug 2026)

ModelRoleOpenTryOn fit
Nemotron 3 Nano 30B-A3BText agents, 1M contextLow (no media)
Nemotron 3 Super 120B-A12BLarger text agentsLow
Nemotron 3 Ultra 550B-A55BFlagship textLow
Nemotron 3.5 Lightning 30B-A3BFast text agentsLow
Nemotron 3 Nano Omni 30B-A3BImage / video / speech / textHigh — Wave 1
Nemotron Voicechat / ASR / Magpie TTSSpeechtts service now exists (ElevenLabs Eleven v4/v4 Turbo shipped 21 Sep 2026) — re-evaluate as an NVIDIA-hosted alternative/local option
Llama-Nemotron embed/rerank VLRAG embeddingsOut of scope

Wave 2 — Virtual try-on (fashion / D2C / marketplace)​

Compile-only as of 29 Aug 2026 — do not integrate until asked. Developers building fitting rooms, PDP/catalog on-model shots, and marketplace listing tools need dedicated person+garment try-on, not another general I2I compose.

Trust order for Path A: hyperscalers and durable public platforms first (Google, Amazon, Alibaba, BFL, Kuaishou, Meta). Fashion specialists we already ship (FASHN, Pruna) stay. Smaller photo-API vendors are watch unless a customer names them. Do not add Fal / Replicate / PiAPI wrappers of models we already call first-party.

NVIDIA has no dedicated VTON NIM. Product roadmap Slice D is still: pick one local OSS path for v0.1.0.

Already in OpenTryOn (do not re-add)​

Registry idVendorKindNotes
flux-vtoBlack Forest LabsDedicated VTON APIFirst-party FLUX VTO
google-vtonGoogle Cloud VertexDedicated VTON APIvirtual-try-on-001; ADC + GOOGLE_CLOUD_PROJECT, not GEMINI_API_KEY
outfitanyone-plusAlibaba Cloud Model StudioDedicated VTON APIaitryon-plus; Beijing-region DASHSCOPE_API_KEY, not Qwen-Image compose
photoroom-vton / photoroom-virtual-modelPhotoroomDedicated VTON + catalog APIImage Editing /v2/edit; shopper try-on or flat-lay → on-model
nova-canvasAmazon BedrockDedicated VTON APIGarment classes incl. footwear
kling-aiKuaishou (Kling / Kolors)Dedicated VTON APIFirst-party Kolors v1 / v1.5
fashn-tryon-max / fashn-tryon-v1.6FASHNDedicated VTON APIFashion suite; v1.6 is the fast e-comm path
p-image-tryonPrunaDedicated VTON APIMulti-garment (up to 11 refs)
segmindSegmindHosted try-on diffusionThird-party hoster; keep, do not add more hosters
nano-banana-2-liteGoogle GeminiComposition I2INot Vertex virtual-try-on-001
qwen-image / qwen-image-localAlibaba QwenComposition I2INot OutfitAnyone aitryon-plus
muse-imageMetaComposition I2IMulti-ref edit, not a garment-fit model

Path A — dedicated VTON APIs to add​

Prefer first-party APIs. Related catalog jobs (product→model, model-swap, parsing) are listed only when they are that vendor’s try-on product.

CandidateVendor durabilityTaskSuggested idWhyStatus
Google virtual-try-on-001Google Cloud (GA 20 Jan 2026; listed discontinue 20 Jan 2027)Shopper / catalog image VTONgoogle-vtonDedicated Vertex / Gemini Enterprise predict API. Person + product image, 1–4 samples, C2PA watermark. Auth is ADC / GCP project, not GEMINI_API_KEY. Distinct from Nano Banana compose. Docsshipped
Alibaba OutfitAnyone-Plus (aitryon-plus)Alibaba Cloud Model StudioImage VTON + combo top/bottomoutfitanyone-plusDedicated DashScope try-on (async). Top, bottoms, dress, face restore, parsing companion aitryon-parsing-v1. Same company as Qwen; Beijing-region key, not the Qwen-Image compose path. Docsshipped
Photoroom Virtual Try-On / Virtual ModelPhotoroom (widely used e-comm photo API)Fitting room or garment→lifestyle modelphotoroom-vton / photoroom-virtual-modelAPI-first catalog/shopper flows. Virtual Model is “flat-lay in, on-model out” (no person photo). Complements dedicated person+SKU VTON. API · Productshipped
Pixelcut Try-OnPixelcut (e-comm photo; Shopify-heavy)Image VTON + garment transferpixelcut-vtonREST /v1/try-on; upper/lower/full. Smaller than Google/Alibaba. APIwatch
FitroomSpecialist startupCombo top+bottom in one requestfitroomStrong e-comm DX; weaker long-term vendor signal.watch
ClaidSpecialistCatalog try-on—Photo-API vendor; overlap with Photoroom.watch
BytePlus Effects / live AR try-onByteDanceReal-time AR, not still VTON—SDK/effects stack. Seedream image/edit is already in the registry (seedream).watch
Tencent Cloud FitDiT (hosted)TencentCommercial FitDiT—Open weights are NC; Tencent Cloud is the commercial door. Confirm a public REST API before Path A.watch
Adobe Firefly ServicesAdobeGeneral gen/edit—Commercially durable Creative Cloud; no dedicated person+garment VTON API found.skip
Shopify / Google Shopping / Walmart ZeekitPlatform lock-inIn-app try-on—Not a developer API we can register.skip
Kling via Fal / PiAPI / ReplicateAggregatorsSame Kolors VTON—We already have first-party kling-ai.skip
Snap Camera Kit / glasses ARSnapAccessory AR—Different modality (mesh/AR), not image VTON.watch

Related try-on jobs (same developers, not a second vton clone):

JobWhat they callPrefer
Shopper fitting roomPerson selfie + SKU photoGoogle VTO, FASHN, Kling, BFL, Amazon, OutfitAnyone
Catalog on-modelFlat-lay → generated modelPhotoroom Virtual Model; FASHN product-to-model (vendor suite — do not invent a parallel adapter until asked)
Multi-SKU outfitTop + bottoms one callOutfitAnyone combo, Fitroom combo, Pruna multi-ref (shipped)
Model swap / consistent modelFace/body swap, keep garmentFASHN model-swap (vendor suite)
Video try-onTemporal garment on a clipCatV2TON local; no durable first-party video-VTON API picked yet
Parsing / hotspotsGarment masks, bboxesAlibaba aitryon-parsing-v1; OpenTryOn already has preprocess helpers

Path B — local / open-weight VTON​

Pick one for v0.1.0 Slice D (tryon.models + opentryon[local]). Many research checkpoints are CC BY-NC-SA — fine for OSS demos, a problem for D2C/marketplace production. Confirm license before making one the default.

CandidateOriginSuggested idVRAM / notesLicense (typical)Status
LeffaCVPR 2025; HF franciszzj/LeffaleffaDiffusers; VITON-HD + DressCode try-on + pose transfer; strong detail/logo storyCode MIT; confirm weight card for commercial D2Cshipped
CatVTON + CatVTON-FLUX LoRAICLR 2025; FLUX.1-Fill LoRA ~37Mcatvton<8GB @ 1024×768; SD 1.5 concatenation pipeline (FLUX LoRA weights exist; official FLUX infer code not released)CC BY-NC-SA 4.0 (code + checkpoints); FLUX-Fill base has its own termsshipped
IDM-VTONECCV 2024; yisol/IDM-VTONidm-vtonHigher fidelity; ~18–24GB typicalCC BY-NC-SAwatch
OOTDiffusionlevihsu/OOTDiffusionootdiffusionCommunity baseline; setup scripts already under tryon/Check repowatch
FitDiTTencent-affiliated DiT; BoyuanJiang/FitDiTfitditHigh garment-detail DiT; ComfyUI existsCC BY-NC-SA; commercial via Tencent Cloudwatch
CatV2TONSame lab as CatVTONcatv2tonVideo try-on; needs a video-VTON design passCheck repo (likely NC like CatVTON)watch
FLUX-fill LoRA (train slice)BFL Fill + brand LoRA—Not a third cloud VTON; opentryon train pathFLUX termswatch
OutfitAnyone weightsHumanAIGC / Alibaba paper—Demos lock person upload; use aitryon-plus API insteadRestricted demosskip
VITON-HD / StableVITON2022–2023 warping/diffusion—Superseded for new workMixedskip
Qwen-Image-Edit-2511 localAlready qwen-image-local—Composition I2I, not a VTON specialist—shipped

By capability (NVIDIA + fashion-relevant others)​

Text-to-image​

CandidatePathNotesStatus
SANA-Sprint (NVLabs, HF Diffusers)BFast local T2I (1–4 steps). Not a NIM.shipped (sana-sprint)
Cosmos 3 T2IB / vLLM-OmniNIM Generator does not expose T2I (one-frame video only in TRT-LLM).watch
FLUX.1-dev / schnell / Kontext via NIMAWe already have BFL FLUX.2. Do not add a second FLUX stack.skip
FLUX.2 Klein 4B via NIMADistilled FLUX.2; only if BFL does not offer Klein.watch
SD 3.5 Large via NIMACrowded T2I table; low fashion differentiation.skip
Qwen-Image via NIMAAlready qwen-image (DashScope) + local.skip
NVIDIA Edify (Getty/Shutterstock)—NIM preview retired 6 June 2025.skip
Janus-Pro-7B (DeepSeek)B (HF, single 24GB GPU)Unified understand + T2I, autoregressive. Only DeepSeek model that generates images; quality trails FLUX.2/Nano Banana/Qwen-Image — OSS option, not a differentiator. Third-party host: Fal if Path A is preferred over local. HFwatch

Image-to-image / edit​

CandidatePathNotesStatus
FLUX.1 Kontext via NIMAIn-context edit. Skip unless we want NIM as a fallback host.skip
Cosmos Transfer 2.5 2BA NIMVideo control transfer (edge/depth/seg/vis), not still I2I. See video.—
SANA-Sprint ControlNetBLocal realtime I2I if we take SANA.watch

Virtual try-on​

NVIDIA / Nemotron has no VTON NIM. Cloud dedicated VTON is already broad (FLUX VTO, Google Vertex, OutfitAnyone-Plus, Photoroom, Amazon Nova, Kling, FASHN, Pruna). Local weights shipped: Leffa (leffa, MIT code) and CatVTON (catvton, CC BY-NC-SA). Full tables: Wave 2 above.

Text-to-video​

CandidatePathNotesStatus
Cosmos 3 Generator nanoAWave 1 shipped (cosmos3). Physics-aware; fashion lookbooks are a stretch but the API is clean.shipped
Cosmos Predict 2.5ALegacy WFM.watch
Muse Video—Consumer preview only; no Meta Model API / weights.blocked

Audio-to-video / talking head​

New CLI service (e.g. audio-to-video) — do not stuff this into video-generate without a design pass.

CandidatePathNotesStatus
NVIDIA LipSync NIMAAudio → lip-dubbed video. Self-hosted gRPC (no hosted REST).shipped (nvidia-lipsync on audio-to-video)
Cosmos 3 + soundB / vLLM-OmniT2V/I2V with synchronized audio. Not on Generator NIM.watch
Pruna P-Video-Avatar—Already shipped (p-video-avatar).shipped

Text-to-speech​

The tts service now exists (shipped 21 Sep 2026 with ElevenLabs) — new candidates go in this table, not a "new CLI service" note.

CandidatePathSuggested idNotesStatus
ElevenLabs Eleven v4A — first-party, synchronous RESTeleven-v4Most expressive TTS; 90+ languages, 10K char limit. ELEVENLABS_API_KEY.shipped
ElevenLabs Eleven v4 TurboA — first-party, synchronous RESTeleven-v4-turbo~100ms latency; audio tags ([whispers], [laughs]) for delivery control. Same key.shipped
NVIDIA Magpie TTS / VoicechatA NIM—Re-evaluate as an NVIDIA-hosted alternative or local option now that tts exists.watch
Qwen-TTS (DashScope)A—Same DASHSCOPE_API_KEY family as Qwen3.8/Wan. Only if a customer asks for a second TTS vendor.watch
OpenAI TTS (gpt-4o-mini-tts etc.)A—Same OPENAI_API_KEY as GPT Image. Only if asked.watch
Local open-weight TTS (e.g. Kokoro, F5-TTS)B—No local TTS shipped yet; worth a Path B pass if someone wants offline/no-per-request-cost voiceover.watch

Image & video understanding / multimodal​

| Gemini 4 Argon (gemini-4-argon) | A — GEMINI_API_KEY, understand | Flagship reasoning multimodal (text/image/video/audio/PDF → text); agentic video + long decode. Plan: Gemini 4 Argon integration plan. Keep shipped gemini-3.8-flash for fast Flash work. | next |

CandidatePathNotesStatus
Gemini 4 Argon (gemini-4-argon)A — GEMINI_API_KEY, understandFlagship reasoning, agentic video, long decode + background Interactions. Plan: gemini-4-argon-integration-plan.shipped
Nemotron 3 Nano OmniAWave 1 shipped. Native audio (unlike current Kimi/Qwen understand tools).shipped
Cosmos 3 Reasoner / Reason2 8BAReasoner shipped (cosmos3-reasoner). Reason2 remains watch.shipped / watch
Muse Glimmer 30B (on NIM)AMeta multimodal on NVIDIA’s catalog; we already have first-party Muse Image. Glimmer is understand, not gen.watch
Llama 3.2 11B/90B Vision on NIMAOlder VLMs; Omni supersedes for new work.skip
DeepSeek-V4.1-Flash (deepseek-flash)A — first-party api-docs.deepseek.com, OpenAI-compatibleCheapest frontier-class VLM available ($0.15/$0.60 per 1M off-peak). Text+image (no video), 1M context, thinking on by default + reasoning_effort (none/low/high/max). Same shape as kimi-k2.6/qwen3.8-max/glm-5.3-flashx. DEEPSEEK_API_KEY.shipped
DeepSeek-VL2 (deepseek-vl2)B — Transformers + DeepSeek's own deepseek_vl2 package (not on PyPI). No first-party hosted API exists (confirmed: deepseek-flash/deepseek-v4-pro are the only hosted models).Local counterpart to deepseek-flash, same pairing pattern as qwen3.8-max/qwen3.8. Default deepseek-ai/deepseek-vl2-tiny (single-GPU friendly).shipped
DeepSeek-OCR (deepseek-ocr)B — HF weights only, no first-party hosted endpoint (confirmed)"Contexts Optical Compression" — real fashion fit: care labels, size tags, SKU/product-copy sheets. Needs flash-attn; distinct value from general captioning even though it shares the understand service.shipped
DeepSeek-V4-Pro (deepseek-v4-pro)A — first-party, text-only (no image input)Flagship reasoning (R-series folded into V4) but no vision — doesn't fit understand's media-first contract. Better framed as an alternate OPENTRYON_AGENT_LLM_PROVIDER backend for the planner than a registry model, mirroring Nemotron's text-only agents.skip
Kimi-K3 (2.8T MoE)A — Moonshot2.8T total / 104B active MoE, Kimi Delta Attention, 1M context, native multimodal. kimi-k3 is shipped — re-verify Moonshot API model id and caps match this generation; if a distinct endpoint appears, add e.g. kimi-k3-moe.watch
Muse Spark OpenUnderstand / agents (coding + perception)Meta — 816B MoE / 42B active; agent- and coding-oriented; 1M context; native audio/video perception; open weights (distinct from muse-image gen).watch
Ling-3.1-flashUnderstand / reasoningInclusionAI — 560B / 25B active hybrid-linear MoE; 1M context. Needs first-party API or open-weight hosting path.watch
BeamUnderstand / agentsReflection — 501B / 23B active frontier MoE; 1M context; Apache-2.0 open weights. Path A hosted chat shipped (beam). Local serve needs four B300s — not registered.shipped
Nemotron 3.5 Super VLUnderstandSee Wave 1 row — nemotron-3.5-super-vl on NIM (NVIDIA_API_KEY).shipped

3D model generation​

New CLI service (e.g. generate-3d). Roadmap lists 3D VTON under Later.

CandidatePathNotesStatus
Microsoft TRELLIS NIMAText-to-3D and image-to-3D meshes; active NVIDIA 3D NIM (Edify 3D is gone).shipped (trellis on generate-3d)
TRELLIS.2 4BBLocal; Windows-skewed wheels as of 2026 — verify Linux before Path B.watch
Video VTON / 3D VTON—Research; no product API picked.watch

Out of scope for this list​

Biology (AlphaFold, Evo2), CFD, weather, routing, chip sim, protein design — on the NIM catalog, not fashion media. DeepSeek-Prover-V2 (formal theorem proving) and DeepSeek-Coder (code LLM) — no fashion/media surface.

Exception shipped anyway: Limite 1B - Violetto (Paradigma Inc, limite-1b-violetto, Path B local) is a text-only 1B math-reasoning model with the same "no fashion/media surface" profile as DeepSeek-Prover-V2/Coder above — it was integrated on explicit request anyway (29 Sep 2026) as a reference point for a possible future generic model gateway direction for OpenTryOn, not because it fits the current fashion/media roadmap. It's named-model-only: no new planner intent, no capability-default change. If that "generic gateway" direction becomes real, re-evaluate whether text-only reasoning/coding/agent models (Nemotron 3 Nano/Super/Ultra, DeepSeek-Coder, DeepSeek-Prover-V2, etc. — currently skip/out-of-scope above) should move into this backlog properly rather than being one-off exceptions.


Frontier long-context MoE survey — Oct 2026​

Public frontier models with ~1M token context and large MoE footprints (industry round-up, Oct 2026). Fashion fit is understand (lookbooks, long PDP copy, video) and planner/agent backends — not VTON/T2I unless noted.

CandidateVendorScale (total / active)HighlightsPathSuggested idStatus
Kimi-K3Moonshot AI2.8T / 104B (896 experts, 16 active)Kimi Delta Attention; native multimodalAkimi-k3 (shipped) and kimi-k3.1-pro (+ -highspeed, shipped 24 Oct 2026). Open weights promised by 4 Nov 2026 — add a Path B local entry when they land.watch
Muse Spark OpenMeta816B / 42BCoding + agents; Blackwell Ultra training; audio/video inA or Bmuse-spark-openwatch
Ling-3.1-flashInclusionAI560B / 25BHybrid-linear MoE reasoningA (or B if weights land)ling-3.1-flashwatch
BeamReflection501B / 23BOpen frontier MoE; Apache-2.0A (hosted). Local weights need four B300sbeamshipped
Nemotron 3.5 Super VLNVIDIA120B / 12BMultimodal VLM; Mamba–Transformer hybrid; HF + NIMA NIMnemotron-3.5-super-vlshipped

Suggested integrate order within this survey: beam shipped (Path A, understand, Oct 2026). Remaining: muse-spark-open / ling-3.1-flash when API or weights are documented → re-check kimi-k3 against Moonshot's latest model card. nemotron-3.5-super-vl is already in the registry.


Fashion-domain survey — 13 Oct 2026​

Fashion-specific models and APIs found on Hugging Face (model API, sorted by recency and likes, 2026), GitHub topic searches and the web. Licences were read from model cards; benchmark numbers are the authors' own and were not reproduced. "Unverified" means a single secondary source. All six rows in this table have shipped.

Fits today's services​

CandidateService / pathWhyLicence / caveatsStatus
FASHN catalogue endpoints — product-to-model, model-create, model-swap, face-to-model, reframe, edit, image-to-video, background-removegenerate / edit / video-generate / bg-remove, Path A, same FASHN_API_KEYWe only wrapped FASHN's try-on. These cover the catalogue workflow (flat-lay → on-model, swap the model, reframe).Hosted; credits per generation_mode × resolution. Integrated without live calls.shipped (fashn-product-to-model, fashn-model-create, fashn-model-swap, fashn-face-to-model, fashn-reframe, fashn-edit, fashn-image-to-video, fashn-bg-remove)
Fashion embeddings — Marqo-FashionSigLIP, MODA (moda-fashion-distilled, moda-pro-lite), ZooClaw-FashionSigLIP2embed, Path B, open_clip / transformersCatalogue search by text or photo with fashion-tuned encoders.Apache-2.0 / MIT. MODA distilled is image-only here. Benchmarks vendor-reported. MODA Plus multi-view recipe, Matryoshka and fp16 variants not wired.shipped (marqo-fashionsiglip, moda-fashion-distilled, moda-pro-lite, zooclaw-fashionsiglip2)
Fashion Florence (anushreeberlia/fashion-florence)understand, Path BFlorence-2-large LoRA (0.77B) → JSON tags: category, colour, material, style. Cheap catalogue tagging.MIT, trust_remote_code. Card: 89.5% category, 42.6% material (exact match); the paper (arXiv 2605.09827) reports 94.6% / 63.0% on a different set.shipped (fashion-florence)
prithivMLmods/QIE-2511-Extract-Outfitedit, Path BWorn photo → flat-lay mockup (garment extraction) on Qwen-Image-Edit-2511; fits our qwen-image-local setup.Apache-2.0 LoRA on an Apache-2.0 base. Community LoRA; needs a full-outfit photo.shipped (qie-2511-extract-outfit)
xocialize/tryon-FLUX.2-klein-4B-loravton, Path BTry-on LoRA on FLUX.2 klein 4B — commercially clean stack. Inputs: masked person + top + bottom garment.Apache-2.0 LoRA + Apache-2.0 base. ~273 training tuples, no metrics, weakest on all-dark outfits.shipped (klein-4b-tryon)
Genlook tryon-safety-classifiergating; needs a design choiceGate garments before rendering them onto a shopper photo (ok / revealing / lingerie / adult + swimwear, subject). 94.6% four-level accuracy (their cross-validation).Apache-2.0. Fixed-label classifier — does not fit the decide question schema; needs a small classify service or a fixed-label adapter.shipped (tryon-safety-classifier, new classify service)

Needs a new service (design pass first)​

CandidateWhat it isNotesStatus
segment — garment / human parsingfashn-ai/fashn-human-parser (SegFormer-B4, 18 classes, 244 MB).Inherits the NVIDIA SegFormer licence (terms not confirmed — possibly non-commercial); segformer_b2_clothes is also "other". Find a commercial-clean parser first. Needs a mask output type.watch — queue: new-services-backlog
measure — body measurements / sizingClad API (clad-body-photo, clad-body-questionnaire) + local clad-body-local on datar-psa/clad-body.New measure service; ISO 8559-1 keys in JSON. CLAD_API_KEY at clad.you/developers.shipped (Oct 2026)
detect — e-commerce detectionTrendyol YOLOv9 (84 product classes) and a DINOv2 256-d embedder.CC BY-SA 4.0 share-alike; paper in review.watch
3D garments / sewing patternspygarment / GarmentCode (MIT); DressWild and GarmentCrafter (2026).Code / weights for the 2026 papers unverified. Large scope.watch

Watch: licence or maturity blockers​

CandidateWhy it mattersBlocker
OmniTry (shoes, bags, jewellery, accessories try-on)Only open model for non-clothing wearables.Apache-2.0 LoRA on FLUX.1-Fill-dev, which is non-commercial.
Mobile-VTON (CVPR 2026), CatV2TON (video try-on)Mobile and video try-on.CC BY-NC-SA / CC BY-NC-ND.
fal/flux-klein-9b-virtual-tryon-lora, fal/virtual-tryoff-loraMost-liked try-on LoRA (156) and its reverse (try-off).Base FLUX.2 klein 9B is non-commercial (~29 GB VRAM); fal-hosted terms not checked.
Garments2Look LoRA (outfit-level try-on with accessories, CVPR 2026)Multi-garment looks.Gated, no licence stated; Qwen bases.
Qwen-Image-2.1 outfit-swap LoRA, Krea-2 outfit-transfer LoRAPopular outfit transfer.Research-licensed base / no licence stated.
FitVTON (MIT, fit-aware), ReGra-VTONFit-aware try-on.Empty README / no model card.
Eevee (CVPR 2026), LiveVVTVideo try-on research.Code and weights unconfirmed.
Outfit compatibility (Polyvore baselines)Styling recommendations.No strong open model; Polyvore is gated.
Hosted vendors: The New Black (API + MCP), Claid, Uwear, LalalandProduct-to-model alternatives.Evidence is vendor blogs only; prefer first-party FASHN for now.

Suggested integration order​

  1. nemotron-omni / cosmos3 / cosmos3-reasoner shipped (Path A, 29 Aug 2026).

  2. google-vton shipped (Path A Vertex virtual-try-on-001, 29 Aug 2026).

  3. outfitanyone-plus / photoroom-vton / photoroom-virtual-model shipped (Path A, 29 Aug 2026).

  4. leffa / catvton shipped (Path B local VTON, 1 Sep 2026).

  5. New services LipSync (A2V), TRELLIS (3D) shipped (audio-to-video / generate-3d, Oct 2026).

  6. SANA-Sprint shipped (sana-sprint on generate, local Diffusers).

  7. Optional: nemotron-omni-local if someone will run 30B-A3B.

  8. deepseek-flash shipped (Path A, understand, 21 Sep 2026).

  9. deepseek-vl2 / deepseek-ocr shipped (Path B local, 21 Sep 2026). Both confirmed no first-party hosted API exists for either.

  10. eleven-v4 / eleven-v4-turbo shipped (Path A, new tts service, 28 Sep 2026). First non-media-generation registry category — new output_kind="audio_bytes" in tryon/cli/runner.py. No dedicated Studio capability screen yet (tracked separately).

  11. fal-h3-max-relight shipped (Path A Fal, video-generate, 8 Oct 2026). Video + lighting-sphere image in, relit video out; no prompt field.

  12. claude-haiku-5-5 shipped (Path A, understand, 8 Oct 2026; model released 7 Oct 2026). First Anthropic adapter in tryon/api/claude.

  13. jev / d1-3b / d1-omni-600m shipped (new decide service, 8 Oct 2026). "System One" decision models: typed questions in, calibrated probabilities out. Jev is API (early access, sign-ups paused 2026-09-22) and text-only; Liquid d1 is Path B local with images (and speech for omni). 13b. videojev-4b / videojev-35b-a3b shipped (Path B local decide, Oct 2026). MCG-NJU VideoJev — video grounding + MC QA; no hosted API (Typesafe Jev/THX remain text-only). Native JSON question array, not Jev noul/choice/score maps.

  14. embeddinggemma-2 / pplx-embed-v2-late-0.6b / pplx-embed-v2-late-9b shipped (new embed service with output_kind="embeddings", 8 Oct 2026). Both local-only: Perplexity's hosted Embeddings API serves the v1 dense models, not v2-late. Hosted counterpart gemini-embedding-2 shipped 13 Oct 2026 (stable id, GEMINI_API_KEY, text/image/audio/video/PDF, --aggregate).

  15. pplx-embed-v2-context-9b-preview — watch: text-only contextual preview, weights/interface may change before GA.

  16. NVIDIA PivotOPD — watch, not integratable today. It is a training method (on-policy distillation for multi-turn agents, arXiv 2609.40285), not a model: the project page lists code as "coming soon" and releases no checkpoints or API. Nothing to wrap in the registry; revisit if NVIDIA publishes student checkpoints.

  17. nano-banana-2.1 shipped (Path A, generate + edit, 8 Oct 2026). Successor to nano-banana-2, which Google deprecated.

  18. fashn-vton-1.5 shipped (Path B local VTON, 8 Oct 2026). Apache-2.0 code + weights; first commercial-friendly local try-on. VRAM not published by FASHN — measure on first GPU run.

  19. gemini-3.8-flash shipped (Path A, understand) and 20. gemini-3.8-flash-tts / gemini-3.8-flash-lite-tts shipped (Path A, second tts vendor), 8 Oct 2026. Voices endpoint / voice design / replication not wired yet. 19b. gemini-4-argon shipped (Path A, understand, Oct 2026): multimodal parity with 3.8 Flash, --video-processing agentic, long decode (max_tokens > 64K + SDK continuation), --background Interactions for text-only jobs. Image/video generation stay on existing nano-banana-* / gemini-omni; video-to-image on Nano Banana 2 / 2.1 / 2 Lite via generate --video. 19c. gemini-4-argon-cyber-permissive shipped (Path A, understand, allowlist): Fairwind defensive-cyber twin of Argon.

Next candidates from the 8 Oct 2026 survey (not started): Hy Image 3.5 Preview (TokenHub, generate/edit), local TTS (Tencent AuK, Voxtral, Breeze 2), Lyria 3.5 (needs a music service).

Oct 2026 batch: kling-v4 shipped (Path A video, 2026 REST API). nemotron-3.5-embed-1b + nemotron-parse shipped (NIM hosted + local NIM twins). flint-local / flint-openrouter and sage-a1.5-local shipped (text understand, local SGLang/vLLM or OpenRouter). Laplace 3B — watch only (laplace-3b-watch; no public weights/API). GAIA-2 — watch only (gaia-2-watch; Wayve latent-diffusion driving world model, no public weights/API). Laguna XS.2 / M.1 shipped (laguna-xs.2, laguna-xs.2.1, laguna-m.1). Arcee Trinity Large shipped (trinity-large 400B Preview, trinity-large-thinking, trinity-mini, trinity-nano local). Recraft V3 shipped (recraft-v3 generate/edit/bg-remove). Ideogram 4.0 was already ideogram.

  1. beam shipped (Path A, understand, Oct 2026): Reflection Beam-1.0-501B-A23B, text chat, REFLECTION_API_KEY. Local four-B300 serve not registered. watch: Muse Spark Open, Ling-3.1-flash, Kimi-K3 2.8T (vs shipped kimi-k3). See Frontier long-context MoE survey.

  2. FASHN catalogue endpoints shipped (Path A, 13 Oct 2026): fashn-product-to-model, fashn-model-create, fashn-model-swap, fashn-face-to-model, fashn-reframe, fashn-edit, fashn-image-to-video, fashn-bg-remove via tryon.api.fashn.FashnAdapter.

  3. Fashion embeddings shipped (Path B, 13 Oct 2026): marqo-fashionsiglip, moda-fashion-distilled, moda-pro-lite, zooclaw-fashionsiglip2 in the embed service.

  4. fashion-florence / qie-2511-extract-outfit shipped (Path B local, understand / edit, 13 Oct 2026).

  5. klein-4b-tryon / tryon-safety-classifier shipped (Path B local, vton / new classify service, 13 Oct 2026).

  6. measure (Clad body photo / questionnaire / local) shipped (new measure service, Oct 2026). Remaining new-service ideas: segment, detect, 3D garments; plus licence-blocked models.

  7. gemini-omni-1.2-flash shipped (6 Nov 2026): default for gemini-omni, adds --duration 3-15s; 1.1 stays selectable via --omni-model (Google routes it to 1.2 from 3 Dec 2026).

  8. ripple-65b-openrouter / ripple-65b-local shipped (6 Nov 2026, understand, text-only): Ricursive Ripple-65B Preview, Apache-2.0. No first-party API exists, so the hosted path is OpenRouter and the local path is a client for your own vLLM/SGLang server. Outside the fashion/media scope, named-model-only.

  9. Inkling 2 Small (Thinking Machines, 594B MoE / 17B active, text + image in, text out) — watch. Launched 5 Nov 2026 on Tinker only with closed weights; Apache-2.0 weights promised after safety testing (reports say roughly the week of 9 Nov). Tinker's OpenAI-compatible endpoint is beta and its docs show checkpoint paths as the model value, serverless sampling is waitlisted, and image input over that endpoint is undocumented. Integrate as Path B (local) when weights publish, and a Path A entry once the hosted endpoint is confirmed. Context length is also inconsistent (Tinker 64K/256K vs model card 1M).

  10. Hima ASR (Prahaladha/hima-asr, ~0.46B, speech-to-text for 11 Indian languages + English, plus a 17.5M "decision head") — needs design. OpenTryOn has no speech-to-text service; this needs a new transcribe service (audio in, text out) and the decision head overlaps decide. The PyTorch path requires trust_remote_code, the licence is Apache-2.0 plus the Gemma Terms of Use, and benchmarks are author-reported. Fashion relevance is limited to voice commerce and support calls.

  11. VFlash (NVIDIA, visual planning for robots and driving) — watch, not integratable: the project page links no VFlash weights, code or API (its "Code" button points at the unrelated LocateAnything repo). Re-check for a release.

  12. Decision-model wave shipped (11 Nov 2026, decide): pplx-decider + pplx-decider-local (Perplexity, first-party Decisions API), clef + clef-local (Cloudflare Workers AI), kev-openrouter + kev-local (Jared Palmer; no first-party API, so OpenRouter), autotrust-jev-local (AutoTrust; no hosted API anywhere). Already present: jev, d1-*, d3-*. watch: Kev 9B/27B on OpenRouter, a hosted AutoTrust endpoint, open weights for Decider v1.2.

  13. Seedream 5.5 (ByteDance, launched 9 Nov 2026) — watch, pending a first-party API id. It is live in Dreamina, CapCut, Doubao and Jimeng, but the BytePlus ModelArk model list (updated 9 Nov) and Volcano Ark list only 5.0 and older. Add it to SEEDREAM_MODELS and the seedream --model-version choices once ModelArk publishes the id. Meanwhile seedream gained 5.0 Flash and corrected 5.0 ids (11 Nov 2026).

  14. lightonocr-3-local / lightonocr-3-vllm shipped (11 Nov 2026, parse): LightOnOCR-3 0.8B / 4B, Apache-2.0, local only (no hosted API). The 1B sibling (LightOnOCR-2 architecture) is not wired.

Follow-ups for the new services: Studio capability screens for decide (question builder + probability bars) and embed (upload + similarity), and a vector-store hand-off for embed output.


Shipped (do not re-add)​

Invoke-layer highlights already in the registry: FLUX.2 (+ Turbo local), FLUX 3 Image (flux3-image), FLUX 3 Image Klein (flux3-image-klein, flux3-image-klein-local), Nano Banana family, GPT Image (1.5 + ChatGPT Images 2.5 Flare/Sunburst), Muse Image, Ideogram 4.0, P-Image-Ideogram, Qwen-Image API+local, Veo 3.1, Sora (deprecated — API removed 2026-09-24), LTX-2.5, Hailuo 2.3, MiniMax H3 / H3 Max, Fal H3 Max (+ Fal H3 Max Lip Sync, fal-h3-max-lipsync), Wan, Runway Gen-4.5, Nemotron Omni, Cosmos 3 Reasoner, Cosmos 3 Generator (+ Cosmos 3 Nano local, cosmos3-local), Kimi K2.6/K2.7/K3, Qwen3.8 (+ Qwen3.8-Omni-Flash, qwen3.8-omni-flash), Hy4 preview (hy4-preview TokenHub + hy4-preview-local vLLM/SGLang), Reflection Beam (beam, text chat), GLM-5.3-FlashX (glm-5.3-flashx, Zhipu/Z.ai — new tryon.api.zai), Ternary Bonsai 2 27B (ternary-bonsai-2-27b, PrismML local server — new tryon.models.ternary_bonsai), P-Video-2-Pro (p-video-2-pro, Pruna MiniMax H3-based), DeepSeek-V4.1-Flash (deepseek-flash, new tryon.api.deepseek), DeepSeek-VL2 (deepseek-vl2, local — new tryon.models.deepseek_vl2), DeepSeek-OCR (deepseek-ocr, local — new tryon.models.deepseek_ocr), ElevenLabs Eleven v4 / Eleven v4 Turbo (eleven-v4 / eleven-v4-turbo, new tts service + tryon.api.elevenlabs), Claude Haiku 5.5 (claude-haiku-5-5, new tryon.api.claude), H3 Max Relight (fal-h3-max-relight), Typesafe Jev / Liquid d1-3B / d1-omni-600M (new decide service), EmbeddingGemma 2 / pplx-embed-v2-late (new embed service), Limite 1B - Violetto (limite-1b-violetto, Paradigma Inc, local — new tryon.models.limite, out-of-scope exception, see above), BEN2, dedicated cloud VTON (flux-vto, google-vton, outfitanyone-plus, photoroom-vton, photoroom-virtual-model, nova-canvas, kling-ai, FASHN, p-image-tryon, Segmind) plus composition try-on (nano-banana-2-lite, qwen-image, muse-image) and local dedicated VTON (leffa, catvton). Full table: CLI --help / registry.


Maintenance​

When adding a row: vendor, modality, Path A/B, proposed registry id, license/API URL, status.
When shipping: move to Shipped, delete the next row, note the registry id.
Re-survey NVIDIA: build.nvidia.com/models + Cosmos / Nemotron blogs. Do not paste the entire NIM catalog.
Re-survey VTON: Vertex Imagen try-on, DashScope OutfitAnyone, Photoroom/Pixelcut, Hugging Face CatVTON / Leffa / IDM-VTON. Prefer first-party APIs over aggregators.