Model Gallery

11 models from 1 repositories

Filter by type:

Filter by tags:

vllm-omni-wan2.2-t2v

Wan2.2-T2V-A14B via vLLM-Omni - Text-to-video generation model from Wan-AI. Generates high-quality videos from text prompts using a 14B parameter diffusion model.

Repository: localaiLicense: apache-2.0

longcat-video

LongCat-Video served by LocalAI's dedicated CUDA backend. Generates video from a text prompt or a start image. The SDPA attention path works without FlashAttention and is suitable for CUDA 13 ARM64 systems such as DGX Spark. This is a very large checkpoint (roughly 83 GB in Hugging Face storage) and requires Linux with an NVIDIA CUDA GPU plus substantial memory and disk.

Repository: localaiLicense: mit

minimax-h3-fl2va-q4

MiniMax-H3 served by vllm.cpp, LocalAI's own C++ port of vLLM. It generates video AND audio jointly from a text prompt, so a clip comes back as an MP4 with a real soundtrack rather than a silent render: ask for speech in the prompt and the model lip-syncs it. This is the Q4_K_M quantisation of the FL2VA partition, which serves text-to-video (t2va) and first/last-frame conditioning (fl2va). Reference conditioning (ref2va) is a different checkpoint and is refused by this one. Roughly 40 GB of weights across five files, plus the two VAE configs that carry the latent statistics. The default canvas is 1344x768 at 124 frames and 24 fps, about 5.2 seconds. Generation is slow — measured at roughly 176 s per denoise step at that canvas on a 20-SM device, so the 50-step default is a multi-hour job. Muxing the finished frames needs ffmpeg on the host.

Repository: localaiLicense: other

minimax-h3-ref2va-q4

MiniMax-H3 served by vllm.cpp, LocalAI's own C++ port of vLLM. It generates video AND audio jointly, so a clip comes back as an MP4 with a real soundtrack rather than a silent render. This is the Q4_K_M quantisation of the Ref2VA partition, the one that takes REFERENCE conditioning: a reference image, a reference clip, or reference audio, prepended as their own blocks so the subject or style carries into the generated video. For plain text-to-video or first/last-frame conditioning use minimax-h3-fl2va-q4 instead - the two partitions are separate checkpoints and each refuses the other's tasks. Use this Q4_K_M build, NOT the NVFP4 Ref2VA weights: NVFP4 renders a multicolour patch grid, and it took three investigations upstream to establish that the fault is the quantisation rather than the reference path. On Q4_K_M the same code renders coherently. Roughly 40 GB of weights across five files, plus the two VAE configs that carry the latent statistics. The default canvas is 1344x768 at 124 frames and 24 fps, about 5.2 seconds. Generation is slow - roughly 176 s per denoise step at that canvas on a 20-SM device, so the 50-step default is a multi-hour job. Muxing the finished frames needs ffmpeg on the host.

Repository: localaiLicense: other

wan-2.1-t2v-1.3b-ggml

Wan 2.1 T2V 1.3B — text-to-video diffusion model, GGUF-quantized for the stable-diffusion.cpp backend. Generates short (33-frame) 832x480 clips from a text prompt. Cheapest Wan variant, suitable for CPU-offloaded inference with ~10 GB of usable RAM.

Repository: localaiLicense: apache-2.0

ltx-2.3-22b-dev-ggml

LTX-2.3 22B dev - DiT-based audio-video foundation model from Lightricks, GGUF-quantized for the stable-diffusion.cpp backend. Generates synchronized video and audio from a text prompt (T2V), a reference image (I2V), or first/last frame pairs (FLF2V). Uses gemma-3-12b-it as the text encoder and ships dedicated video and audio VAEs plus an embeddings_connectors safetensors that bridges the LLM hidden states to the diffusion model. This entry uses the dynamic (UD) Q4_K_M quantization of the 22B model (~16 GB) paired with the UD-Q4_K_XL QAT Gemma encoder (~7.4 GB). Recommended generation: width=1280, height=720, video_frames=33, fps=24, sampler=euler, cfg_scale=6.0.