# AI OSS Daily · Picks archive · 2026-07-19

Auto-updated daily around 09:00 and pushed straight to your Telegram — free, no ads, unsubscribe anytime.

- [thinkingmachines/Inkling](https://cenxialiu7-cloud.github.io/ai-oss-daily/en/p/model-thinkingmachines-inkling/) — image-text-to-text · transformers, safetensors, inkling_mm_model
- [jdopensource/JoyAI-Echo](https://cenxialiu7-cloud.github.io/ai-oss-daily/en/p/model-jdopensource-joyai-echo/) — text-to-video · ltx-video, JoyAI-Echo, text-to-video
- [stabilityai/stable-audio-3-medium](https://cenxialiu7-cloud.github.io/ai-oss-daily/en/p/model-stabilityai-stable-audio-3-medium/) — text-to-audio · stable-audio-3, safetensors, audio-generation
- [OpenMOSS-Team/MOSS-TTS-v1.5](https://cenxialiu7-cloud.github.io/ai-oss-daily/en/p/model-openmoss-team-moss-tts-v1-5/) — text-to-speech · safetensors, moss_tts_delay, text-to-speech
- [prism-ml/Ternary-Bonsai-27B-gguf](https://cenxialiu7-cloud.github.io/ai-oss-daily/en/p/model-prism-ml-ternary-bonsai-27b-gguf/) — text-generation · llama.cpp, gguf, conversational
- [nineninesix/gepard-1.0](https://cenxialiu7-cloud.github.io/ai-oss-daily/en/p/model-nineninesix-gepard-1-0/) — text-to-speech · transformers, safetensors, qwen3_5_text
- [Wan-AI/Wan-Dancer-14B](https://cenxialiu7-cloud.github.io/ai-oss-daily/en/p/model-wan-ai-wan-dancer-14b/) — image-to-video · diffusers, safetensors, i2v
- [ostris/krea2_turbo_style_reference](https://cenxialiu7-cloud.github.io/ai-oss-daily/en/p/model-ostris-krea2-turbo-style-reference/) — text-to-image · diffusers, text-to-image, lora
- [AlperKTS/Krea2_FP8](https://cenxialiu7-cloud.github.io/ai-oss-daily/en/p/model-alperkts-krea2-fp8/) — text-to-image · diffusers, krea2, text-to-image
- [Supertone/supertonic-3](https://cenxialiu7-cloud.github.io/ai-oss-daily/en/p/model-supertone-supertonic-3/) — text-to-speech · supertonic, onnx, text-to-speech
- [google/magenta-realtime-2](https://cenxialiu7-cloud.github.io/ai-oss-daily/en/p/model-google-magenta-realtime-2/) — text-to-audio · magenta-realtime-2, tflite, text-to-audio
- [nvidia/Cosmos3-Super-Image2Video](https://cenxialiu7-cloud.github.io/ai-oss-daily/en/p/model-nvidia-cosmos3-super-image2video/) — image-to-video · cosmos, diffusers, safetensors