Vision Model**](https://huggingface.co/stepfun-ai/Step-3.7-Flash) - Delivers a vision model designed for local AI agents.
2. [**NVIDIA LocateAnything-3B**](https://huggingface.co/nvidia/LocateAnything-3B) - Delivers one-step visual grounding for object detection.
3. [**Qwen3.6-35B-A3B-Uncensored-Genesis-V2-APEX-MTP-GGUF**](https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-V2-APEX-MTP-GGUF/) - Removes all refusals from the model.
4. [**Qwen3.5-27B-uncensored-heretic-v2-Native-MTP-Preserved**](https://huggingface.co/llmfan46/Qwen3.5-27B-uncensored-heretic-v2-Native-MTP-Preserved) - Removes 89% of AI refusals.
5. [**Keye-VL-2.0-30B-A3B**](https://github.com/Kwai-Keye/Keye) - Brings native agent tools to long video AI.
6. [**NuExtract3**](https://huggingface.co/numind/NuExtract3) - Turns sensitive docs into markdown without the cloud.
7. [**Gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic**](https://huggingface.co/llmfan46/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic) - An uncensored creative wordsmith model.
8. [**Gemma4-26B-A4B-Uncensored-HauhauCS-Balanced**](https://huggingface.co/HauhauCS/Gemma4-26B-A4B-Uncensored-HauhauCS-Balanced) - Drops with zero refusals for balanced chats.
9. [**Intern-S2-Preview**](https://huggingface.co/internlm/Intern-S2-Preview) - Packs trillion-scale science smarts into a 35B model.
10. [**Fara-7B**](https://huggingface.co/microsoft/Fara-7B) - A tiny AI agent that runs your web chores privately.
11. [**Marlin-2B**](https://huggingface.co/NemoStation/Marlin-2B) - Pins down every second of your video.
12. [**Qwopus3.5-9B-Coder-GGUF**](https://huggingface.co/Jackrong/Qwopus3.5-9B-Coder-GGUF) - Puts a private coding agent on your laptop.
13. [**Lance**](https://github.com/bytedance/Lance) - Unifies image and video generation in one lightweight model.
14. [**SenseNova-U1-A3B-MoT**](https://huggingface.co/sensenova/SenseNova-U1-A3B-MoT) - A unified vision-language powerhouse that runs locally.
15. [**Qwopus3.6-27B-v2-MTP-GGUF**](https://huggingface.co/Jackrong/Qwopus3.6-27B-v2-MTP-GGUF) - Puts faster stepwise AI on your GPU.
16. [**Qwen3.6-35B-A3B-uncensored-heretic-Native-MTP-Preserved**](https://huggingface.co/llmfan46/Qwen3.6-35B-A3B-uncensored-heretic-Native-MTP-Preserved) - Delivers fewer refusals for open chat.
17. [**Unsloth Qwen3.6-27B-GGUF-MTP**](https://huggingface.co/unsloth/Qwen3.6-27B-GGUF-MTP) - Drops a model for 2x faster local AI.
18. [**Ovis2.6-80B-A3B**](https://huggingface.co/AIDC-AI/Ovis2.6-80B-A3B) - Lands private visual AI on a single GPU.
19. [**Qwen3.6-27B-MTP-UD-GGUF**](https://huggingface.co/havenoammo/Qwen3.6-27B-MTP-UD-GGUF) - Makes your GPU think ahead for faster outputs.
20. [**MiniCPM-V-4.6**](https://huggingface.co/openbmb/MiniCPM-V-4.6) - Packs private visual AI into phones.
21. [**Qwen3.5-9B-DeepSeek-V4-Flash-GGUF**](https://huggingface.co/Jackrong/Qwen3.5-9B-DeepSeek-V4-Flash-GGUF) - Brings deep reasoning home.
22. [**Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF**](https://huggingface.co/DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF) - A massive uncensored fine-tune for power users.
23. [**Gemma-4-26B-A4B-it-assistant**](https://huggingface.co/google/gemma-4-26B-A4B-it-assistant) - Google turbocharges Gemma 4 for assistant tasks.
24. [**Gemma-4-31B-it-assistant**](https://huggingface.co/google/gemma-4-31B-it-assistant) - Google drops a model to triple local AI speed.
25. [**Nemotron-3-Nano-Omni-30B-A3B-Reasoning-NVFP4**](https://huggingface.co/nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-NVFP4) - Opens local multimodal AI with reasoning.
26. [**Mistral-Medium-3.5-128B**](https://huggingface.co/mistralai/Mistral-Medium-3.5-128B) - Introduces a massive unified tool.

**🖼️ Image**

1. [**Microsoft Lens-Turbo**](https://huggingface.co/microsoft/Lens-Turbo) - Delivers instant 1440p images in four steps flat.
2. [**Microsoft Lens**](https://huggingface.co/microsoft/Lens) - Focuses high-quality image creation on your home GPU.
3. [**Nvidia

PiD**](https://huggingface.co/nvidia/PiD) - Fuses upscaling and decoding for instant 4K images.
4. [**Anima Base v1.0**](https://huggingface.co/circlestone-labs/Anima) - Spawns anime art straight from text prompts.
5. [**HiDream-O1-Image**](https://huggingface.co/HiDream-ai/HiDream-O1-Image) - Crafts multi-task visuals straight from raw pixels.
6. [**Walkyrie-1.3B-v1.0**](https://huggingface.co/kpsss34/Walkyrie-1.3B-v1.0) - Spins video smarts into speedy local image creation.
7. [**UltraReal_FineTune_Anima**](https://huggingface.co/Danrisi/UltraReal_FineTune_Anima) - Delivers analog soul to digital photos.

**🎬 Video**

1. [**LongCat-Video-Avatar-1.5**](https://huggingface.co/meituan-longcat/LongCat-Video-Avatar-1.5) - Materializes studio-quality talking heads locally.
2. [**SANA-WM Bidirectional**](https://huggingface.co/Efficient-Large-Model/SANA-WM_bidirectional) - Turns one still image into a minute-long 3D video.
3. [**Causal-Forcing**](https://github.com/thu-ml/Causal-Forcing) - Distills real-time video generation for a single GPU.
4. [**LTX2.3-10Eros**](https://huggingface.co/TenStrip/LTX2.3-10Eros) - Brings still images to life with layered precision.

**🎧 Audio**

1. [**MOSS-TTS-v1.5**](https://huggingface.co/OpenMOSS-Team/MOSS-TTS-v1.5) - Lands with precise pause controls and 31-language synthesis.
2. [**DramaBox**](https://github.com/resemble-ai/DramaBox) - Interprets stage directions for expressive AI voiceovers.
3. [**Supertonic-3**](https://huggingface.co/Supertone/supertonic-3) - Whispers 31 languages directly from your device.
4. [**Scenema-Audio**](https://huggingface.co/ScenemaAI/scenema-audio) - Lets you direct voices with emotion and scene sounds.

**🤖 Agents**

1. [**SmallCode**](https://github.com/Doorman11991/smallcode) - A local coding agent that squeezes power from small LLMs.
2. [**Opendesk**](https://github.com/vitalops/opendesk) - Unlocks direct desktop control for AI agents.

**⚡ LoRA**

1. [**ControlLight**](https://github.com/yfyang007/ControlLight) - Turns photo brightening into a smooth dimmer switch.
2. [**LTX-2.3 Upscale IC Lora**](https://huggingface.co/Zlikwid/LTX_2.3_Upscale_IC_Lora) - Breathes detail into fuzzy video renders.
3. [**VR-360-Outpaint-LTX2.3-IC-LoRA**](https://huggingface.co/TheBurgstall/VR-360-Outpaint-LTX2.3-IC-LoRA) - Morphs clips into 360 VR scenes.
4. [**SYSTMS-FLW-IC-LORA-LTX-2.3**](https://huggingface.co/systms/SYSTMS-FLW-IC-LORA-LTX-2.3) - Smooths shot transitions using a simple gray frame.
5. [**LTX-2.3-Dearchive-Lora**](https://huggingface.co/oumoumad/ltx-2.3-dearchive-lora) - Turns vintage grain into sharp modern video.
6. [**Obscura Remova**](https://huggingface.co/WepeNerd/Obscura_Remova) - Wipes away visual clutter from video scenes.
7. [**OmniNFT LoRA Adapters**](https://github.com/zghhui/OmniNFT) - Fixes lip-sync and audio-video alignment for LTX.
8. [**Ace-Step-1.5-XL-Concept-Sliders**](https://huggingface.co/Xanthius/Ace-Step-1.5-XL-Concept-Sliders) - Dials up fine-grained AI music control.
9. [**LTX-2.3-22b-IC-LoRA-LipDub**](https://huggingface.co/Lightricks/LTX-2.3-22b-IC-LoRA-LipDub) - Magically redubs videos with a text prompt.
10. [**Flux.2-Klein-Loras**](https://huggingface.co/DeverStyle/Flux.2-Klein-Loras) - Summons six style LoRAs for creative edits.
11. [**Qwen-2512-portrait**](https://huggingface.co/a3xrfgb/Qwen-2512-portrait) - Erases the plastic look from AI portraits.

**🏋️ Training**

1. [**IMG-Dataset-Refiner**](https://github.com/NyxAwroo/IMG-Dataset-Refiner) - Scrubs image folders into perfect AI training data.
2. [**Anima-TrainFlow**](https://github.com/ThetaCursed/Anima-TrainFlow) - Corrals LoRA training into one page.
3. [**Bracket**](https://github.com/tlennon-ie/bracket) - Delivers stat-backed winners so you stop guessing configs.

**🛠️ Other Tools**

1. [**hipEngine**](https://github.com/shisa-ai/hipEngine) - Hacks AMD GPUs to run massive AI locally.
2. [**MiniCPM-V-4.6-OrangePi**](https://github.com/lvyufeng/minicpm-v-4.6-orangepi) - Boots full multimodal AI on a sub-100 dollar board.
3. [**ScreenDiffusion

V0.2**](https://github.com/rudyaa-sd/ScreenDiffusion) - Reimagines your live screen as evolving artwork.
4. [**OpenReader**](https://github.com/richardr1126/openreader) - Preloads audio to make every document a private audiobook.
5. [**FP-Background_Obliterator**](https://github.com/frozenpepper/FP-Background_Obliterator) - Carves out local AI cutouts for your images.
6. [**Streamlined-HF-Model-Search**](https://github.com/stew675/streamlined-hf-model-search) - Makes local AI model discovery a breeze.
7. [**Studiomi300**](https://github.com/bladedevoff/studiomi300) - Spins one prompt into a 30s cinematic reel.
8. [**MusiCue**](https://github.com/cedarconnor/MusiCue) - Chisels music into frame-perfect animation cues.
9. [**Tokenspeed**](https://github.com/MikeVeerman/tokenspeed/) - Streams fake tokens to let you feel LLM speed.
10. [**ExLlamaV3**](https://github.com/turboderp-org/exllamav3) - Supercharges home AI with triple-speed DFlash decoding.
11. [**Needle**](https://github.com/cactus-compute/needle) - Stitches seamless tool calling for budget phones.
12. [**Lucebox-Hub**](https://github.com/Luce-Org/lucebox-hub) - Supercharges AMD Strix Halo with DFlash and PFlash.
13. [**Derpy-Turtle-The-Kokoro-Trainer**](https://github.com/BovineOverlord/Derpy-Turtle-The-Kokoro-Trainer) - Hatches smooth voice clones locally.
14. [**TextGen Portable**](https://github.com/oobabooga/textgen) - Run AI locally with no install and no telemetry.
15. [**Merlin-Community**](https://github.com/corbenicai/merlin-community/) - Drops redundant AI words for leaner conversations.
16. [**AI Metadata Viewer**](https://github.com/GChenSi-2/ai-metadata-viewer) - Reveals hidden prompts inside any AI image.
17. [**Cull**](https://github.com/tlennon-ie/cull) - Slashes AI image sorting time without cloud dependencies.
18. [**FP16-FP8-to-NVFP4**](https://github.com/thenotrealuser/fp16-fp8-to-nvfp4) - Trims 31GB AI models down to 13GB.
19. [**ShrinkComfy**](https://github.com/Virgile-fr/ShrinkComfy) - Shrinks your ComfyUI PNGs without erasing workflow data.
20. [**EasyUI**](https://github.com/kigy1/EasyUI) - Turns messy AI node graphs into simple user interfaces.
21. [**Torch-Nvenc-Compress**](https://github.com/shootthesound/torch-nvenc-compress) - Turns idle video chips into AI data superchargers.
22. [**Deepbooru-Tagwalker**](https://github.com/Elliezrah/deepbooru-tagwalker) - Walks tags first to simplify dataset verification.
23. [**Diff-forge**](https://github.com/Oqura-ai/diff-forge) - Carves flawless training datasets from your video footage.
24. [**Caption-Creator**](https://github.com/Merserk/Caption-Creator) - Lands local image captioning that skips the cloud.
25. [**Ace-Step-1.5-Api-server-UI**](https://github.com/tritant/Ace-Step-1.5-Api-server-UI) - Turns your browser into a local AI music studio.
26. [**Phosphene**](https://github.com/mrbizarro/phosphene) - Stitches visuals and sound instantly on Macs.
27. [**Beellama.cpp**](https://github.com/Anbeeld/beellama.cpp) - Supercharges local AI with a speed overhaul.
28. [**ds4.pinokio**](https://github.com/cocktailpeanut/ds4.pinokio) - Slots a full DeepSeek V4 brain into Apple Silicon Macs.

**+ ComfyUI Custom Nodes & Tools**

**ComfyUI Performance & Hardware Tools**
1. [**ComfyUI-FeatherOps**](https://github.com/woct0rdho/ComfyUI-FeatherOps) - Injects AMD RDNA3 GPUs with a massive 50% diffusion speed boost.
2. [**ComfyUI-SPEED**](https://github.com/ruwwww/ComfyUI-SPEED) - Dials up image generation to nearly double the standard speed.
3. [**Comfyui-Mesh**](https://github.com/shootthesound/comfyui-mesh) - Splits large AI models across two GPUs without needing NVLink.
4. [**ComfyUI-Safe-Chunked-Image-Blend**](https://github.com/xmarre/ComfyUI-Safe-Chunked-Image-Blend) - Defeats system freezes by handling image blending in chunks.
5. [**BangtrixToolkit**](https://github.com/Anonymzx/BangtrixToolkit) - Gives ComfyUI a live hardware monitor and a prompt translator.

**ComfyUI Video & Audio Workflows**
1. [**Vlo**](https://github.com/PxTicks/vlo/) - Gives creators a timeline to finesse generative AI

footage.
2. [**Comfyui-Controlfoley**](https://github.com/SGUN-father/comfyui-controlfoley) - Syncs AI Foley sound effects directly to your silent footage.
3. [**ComfyUI-DramaBox**](https://github.com/FranckyB/ComfyUI-DramaBox) - Injects expressive AI speech directly into visual workflows.
4. [**Comfyui_VideoCombine_Plus**](https://github.com/peterducan-hub/Comfyui_VideoCombine_Plus) - Brings audio control and frame saving options to video exports.

**ComfyUI Image Editing & Upscaling**
1. [**ComfyUI_KleinTiledUpscaler**](https://github.com/Gavr728/ComfyUI_KleinTiledUpscaler) - Debuts seamless upscaling specifically for Flux2 Klein.
2. [**ComfyUI-PiD**](https://github.com/Merserk/ComfyUI-PiD) - Bypasses the VAE for one-step pixel diffusion upscaling.
3. [**Orion4D_generative_paint**](https://github.com/orion4d/Orion4D_generative_paint) - Brings layered drawing capabilities to ComfyUI workflows.
4. [**ComfyUI-Olm-Liquify**](https://github.com/o-l-l-i/ComfyUI-Olm-Liquify) - Brings liquify-style warping effects to your generations.
5. [**ComfyUI-Angelo**](https://github.com/shootthesound/ComfyUI-Angelo) - Brings "click to fix" editing to AI image generation.
6. [**ComfyUI-PlagueKind-Nodes**](https://github.com/PlagueKind/ComfyUI-PlagueKind-Nodes) - Tames mask drift for flawless inpainting results.
7. [**ComfyUI-Fayens**](https://github.com/iFayens/ComfyUI-Fayens) - Brings cinematic polish to face swap workflows.
8. [**ComfyUI-ReferenceLatentPlus**](https://github.com/shootthesound/comfyui-ReferenceLatentPlus) - Enables per-image strength dialing for finer control.

**ComfyUI Prompting & Model Management**
1. [**ComfyUI-SmartPromptCrafter**](https://github.com/jideka/ComfyUI-SmartPromptCrafter) - Auto-matches prompts to any model you are using.
2. [**RebelsPromptEnhancer**](https://github.com/RealRebelAI/RebelsPromptEnhancer) - Offers a local prompt boost for private workflows.
3. [**ComfyUI-Anima-Style-Nodes**](https://github.com/fulletLab/comfyui-anima-style-nodes) - Brings visual tag selection to ComfyUI.
4. [**Comfyui-Anima-Regional-Conditioning**](https://github.com/Sen-sou/Comfyui-Anima-Regional-Conditioning) - Paints prompts in bounded regions for specific details.
5. [**ComfyUI-lora-FindingLora**](https://github.com/shootthesound/comfyui-lora-FindingLora) - Delivers quick LoRA finding and stacking capabilities.
6. [**ComfyUi-Untwisting-RoPE**](https://github.com/BigStationW/ComfyUi-Untwisting-RoPE) - Delivers Untwisting-RoPE for style transfer without copying.

**ComfyUI Workflow Management & Utilities**
1. [**Nexus-BTA**](https://github.com/JpAndreBTA/Nexus-BTA) - Transforms your setup into a complete local creative studio.
2. [**Somni-ComfyUI**](https://github.com/searcc/somni-comfyui) - Serves up a slick frontend for ComfyUI on any device.
3. [**ComfyUI-Artius-Browser**](https://github.com/AlexYez/comfyui-artius-browser) - Delivers snappy sidebar browsing for creators.
4. [**ComfyUI-Workflow-Finder**](https://github.com/gregowahoo/comfyui-workflow-finder) - Finds ComfyUI workflows just by describing them.
5. [**WorkflowX-Configurator**](https://github.com/haroonaslam/WorkflowX-Configurator) - Lets you switch ComfyUI profiles in just one click.
6. [**Comfyui-Node-Canvas**](https://github.com/caoool/comfyui-node-canvas) - Spins custom ComfyUI nodes from visual blueprints.
7. [**ComfyUI-Magos-Nodes**](https://github.com/MagosDigitalStudio/ComfyUI-Magos-Nodes) - Drops a full skeleton editor right inside ComfyUI.
8. [**Huge ComfyUI-Yedp-Action-Director Update**](https://github.com/yedp123/ComfyUI-Yedp-Action-Director) - Updates the 3D Director in ComfyUI with new features.
9. [**ComfyUI-ialhabbal**](https://github.com/ialhabbal/ComfyUI-ialhabbal) - Bundles eight AI image tools into one convenient node pack.
10. [**ComfyUI_ShowMe**](https://github.com/SKBv0/ComfyUI_ShowMe) - Drops explainable AI sketches right on your workflow.
11. [**ComfyUI-XAV-Google-Sheets**](https://github.com/XAV-Games/ComfyUI-XAV-Google-Sheets) - Pipes spreadsheet text directly into AI workflows.
12.

[**ComfyUI-Clippy-Reloaded**](https://github.com/shootthesound/comfyui-clippy-reloaded) - Pastes clipboard images without needing to save files first.
13. [**ComfyUI-gonztok_nodes**](https://github.com/gonztok/ComfyUI-gonztok_nodes) - Ditches file paths by using visual pickers instead.

**Need to go further back?** Check out [**last month's post**](https://www.reddit.com/r/StableDiffusion/comments/1t0bm5w/local_ai_news_you_missed_april_2026/) or the full archive at [**LocalAI News**](https://localainews.co/news/news-you-missed/). If there's anything wrong, let me know in the comments.

PS: Few days behind on this one (I can already see like 30 releases in the past few days I've missed. If you did an average of releases for the month, its around 5 - 7 daily). There's still a lot to work to be done so plans for this month include optimizing and building better automations, updating the site's design, adding non news related content and more useful tools.

https://redd.it/1ttpi6d
@rStableDiffusion

I compared 62 samplers and 16 schedulers for WAN 2.1 image generation and rated the image quality so you don't have to 😬

https://preview.redd.it/mj7g2e2xtn4h1.png?width=616&format=png&auto=webp&s=5ce6a688fafdee4ce578ad81925f216edb32fe8f

Here's a sampler/scheduler comparison table for WAN 2.1 image generation. Obviously it reads like Red < Orange < Yellow < Green. You're welcome!


https://redd.it/1ttpvn4
@rStableDiffusion

FLUX.2 Klein 9B Schematic LoRA - Depth, Normal, Pose, and Segmentation

https://redd.it/1ttpbl3
@rStableDiffusion

Cosmos3 Nano testing with vllm-omni

https://redd.it/1tttwgu
@rStableDiffusion

"Synchrotron" Audioreactive text2video (Stable Audio 3 + LTX 2.3)
https://youtu.be/VzwiJKgagqQ

https://redd.it/1tu09k1
@rStableDiffusion

Retro Anime trailers with ViduQ3 x GPT-2-img with some sprinkles of seedance2 to get retro anime aestetics. A pain to edit.

https://redd.it/1tu4fzt
@rStableDiffusion

I've been trying to replicate this Anima V1.0 image all day but I can't manage it. Maybe I need a special workflow or something?

https://redd.it/1tu4zsy
@rStableDiffusion

Cosmos3-Super-Image2Video running locally on a single RTX PRO 6000 96GB

https://redd.it/1tu7bth
@rStableDiffusion