vLLM
High-throughput LLM serving for GPUs.
Open Source 24–80 GB VRAM
- Min VRAM
- 24 GB
- GPU class
- Datacenter GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
48 GB VRAM37 tools fit this tier
AI tools that run on 48 GB VRAM
Forty-eight gigabytes is workstation territory. Now you can run the biggest open-weight video models at full precision, serve 70B LLMs through vLLM, and start fine-tuning instead of just LoRA-ing. The RTX 5090, RTX A6000, and L40S are the typical homes for this tier.
Typical GPUs in this tier
NVIDIA RTX 5090NVIDIA RTX A6000NVIDIA L40 / L40S
Capabilities
What you can run
- Wan 2.2 14B at BF16
- HunyuanVideo at full precision
- 70B LLMs in FP8 with vLLM
- Flux full fine-tuning (with gradient checkpointing)
- Multi-model ComfyUI pipelines without offload
What stays out of reach
- Datacenter-only workloads (H100/H200 quirks)
- Truly massive 405B+ LLMs without sharding
Recommended tools for 48 GB VRAM
Sorted by best fit for this tier — tools designed around your VRAM budget first, then by our power-user score.
NVIDIA Cosmos
World-foundation models for physical AI.
Open Source 24–80 GB VRAM
- Min VRAM
- 24 GB
- GPU class
- Datacenter GPU
- Quant
- FP16
Actually FreeWatermark-FreeAPI
HunyuanVideo
13B open-weight cinematic text-to-video.
Open Source 24–48 GB VRAM
- Min VRAM
- 24 GB
- GPU class
- Workstation GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Magi-1
Autoregressive video diffusion at 24 GB.
Open Source 24–48 GB VRAM
- Min VRAM
- 24 GB
- GPU class
- Workstation GPU
- Quant
- FP8
Actually FreeNo SignupOpen SourceWatermark-Free
Mochi 1
Genmo's 10-B open-weight T2V — the first 'genuinely fluid' OSS video model.
Open Source 24–60 GB VRAM
- Min VRAM
- 24 GB
- GPU class
- Workstation GPU
- Quant
- BF16
Actually FreeNo SignupOpen SourceWatermark-Free
Modal
Serverless Python for GPU workloads.
Freemium 16–80 GB VRAM
- Min VRAM
- 16 GB
- GPU class
- Datacenter GPU
- Quant
- —
Actually FreeWatermark-FreeHobbyist-FriendlyAPI
AI-Toolkit (Ostris)
Modern training framework — Flux, SDXL, SD3 LoRAs in YAML.
Open Source 16–24 GB VRAM
- Min VRAM
- 16 GB
- GPU class
- High-end GPU
- Quant
- BF16
Actually FreeNo SignupOpen SourceWatermark-Free
TRELLIS
Microsoft Research's structured 3D representation model.
Open Source 16–24 GB VRAM
- Min VRAM
- 16 GB
- GPU class
- High-end GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Pyramid Flow
Memory-efficient T2V via pyramidal flow matching.
Open Source 16–24 GB VRAM
- Min VRAM
- 16 GB
- GPU class
- High-end GPU
- Quant
- BF16
Actually FreeNo SignupOpen SourceWatermark-Free
Wan 2.2
Open-weight video diffusion from Alibaba.
Open Source 12–48 GB VRAM
- Min VRAM
- 12 GB
- GPU class
- Workstation GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Kohya_ss
The standard SDXL/Flux LoRA training UI.
Open Source 12–24 GB VRAM
- Min VRAM
- 12 GB
- GPU class
- High-end GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
3D Gaussian Splatting
The INRIA original — train your own splats.
Open Source 12–24 GB VRAM
- Min VRAM
- 12 GB
- GPU class
- Mid GPU
- Quant
- FP32
Actually FreeNo SignupWatermark-Free
Hunyuan3D-2
Tencent's open 3D generator — multi-view, PBR, ready-to-use meshes.
Open Source 12–16 GB VRAM
- Min VRAM
- 12 GB
- GPU class
- High-end GPU
- Quant
- FP16
Actually FreeOpen SourceWatermark-FreeAPI
LTX-Video
Real-time-ish open video diffusion from Lightricks.
Open Source 12–16 GB VRAM
- Min VRAM
- 12 GB
- GPU class
- High-end GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
OneTrainer
Modern alternative trainer for SD/SDXL/Flux.
Open Source 12–24 GB VRAM
- Min VRAM
- 12 GB
- GPU class
- High-end GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Stable Diffusion 3.5 Large
Stability's MMDiT flagship at 8B params.
Open Source 12–24 GB VRAM
- Min VRAM
- 12 GB
- GPU class
- Mid GPU
- Quant
- FP16
Actually FreeWatermark-FreeHobbyist-FriendlyAPI
SUPIR
Diffusion-based photorealistic upscaler.
Open Source 12–24 GB VRAM
- Min VRAM
- 12 GB
- GPU class
- Mid GPU
- Quant
- FP16
Actually FreeNo SignupWatermark-FreePlugin
CogVideoX 5B
Open-source text-to-video diffusion from THUDM.
Open Source 12–24 GB VRAM
- Min VRAM
- 12 GB
- GPU class
- High-end GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
FluxGym
Dead-simple Flux LoRA training in a Gradio UI.
Open Source 12–20 GB VRAM
- Min VRAM
- 12 GB
- GPU class
- Mid GPU
- Quant
- FP8
Actually FreeNo SignupOpen SourceWatermark-Free
Stable Video Diffusion
Image-to-video diffusion — 25 frames, 14 or 25 steps.
Open Source 12–16 GB VRAM
- Min VRAM
- 12 GB
- GPU class
- High-end GPU
- Quant
- FP16
Actually FreeOpen SourceWatermark-FreeHobbyist-Friendly
ComfyUI-AnimateDiff-Evolved
Animation motion modules for ComfyUI.
Open Source 10–16 GB VRAM
- Min VRAM
- 10 GB
- GPU class
- High-end GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Diffusers
Hugging Face's go-to library for every diffusion model.
Open Source 4–12 GB VRAM
- Min VRAM
- 4 GB
- GPU class
- Entry GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
AUTOMATIC1111 (stable-diffusion-webui)
The original SD power-user webUI.
Open Source 4–12 GB VRAM
- Min VRAM
- 4 GB
- GPU class
- Mid GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Topaz Video AI
GPU-accelerated upscaling, frame-interp, denoise.
Paid · from $299 4–12 GB VRAM
- Min VRAM
- 4 GB
- GPU class
- Mid GPU
- Quant
- —
Watermark-FreeHobbyist-Friendly
WhisperX
Whisper + speaker diarisation + word-level timestamps.
Open Source 4–8 GB VRAM
- Min VRAM
- 4 GB
- GPU class
- Entry GPU
- Quant
- INT8
Actually FreeOpen SourceWatermark-FreeHobbyist-Friendly
XTTS v2 (Coqui)
Multilingual voice cloning in 6 seconds.
Open Source 4–6 GB VRAM
- Min VRAM
- 4 GB
- GPU class
- Entry GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Stable Diffusion WebUI Forge
Optimized A1111 fork for low-VRAM cards.
Open Source 4–8 GB VRAM
- Min VRAM
- 4 GB
- GPU class
- Entry GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Flowframes
Free RIFE-based frame interpolation.
Free 4–8 GB VRAM
- Min VRAM
- 4 GB
- GPU class
- Entry GPU
- Quant
- —
Actually FreeNo SignupOpen SourceWatermark-Free
Tabby
Self-hosted, GPU-accelerated coding autocompletion.
Freemium · from $19/mo 4–8 GB VRAM
- Min VRAM
- 4 GB
- GPU class
- Entry GPU
- Quant
- GGUF
Actually FreeNo SignupOpen SourceWatermark-Free
RVC (Retrieval-based Voice Conversion)
The voice-changer that took over Discord.
Open Source 4–8 GB VRAM
- Min VRAM
- 4 GB
- GPU class
- Entry GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Fooocus
Stable Diffusion XL, dialed to one button.
Open Source 4–8 GB VRAM
- Min VRAM
- 4 GB
- GPU class
- Entry GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Tortoise TTS
Slow, but the quality is worth the wait.
Open Source 4–8 GB VRAM
- Min VRAM
- 4 GB
- GPU class
- Entry GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Transformers
The library every LLM ships against first.
Open Source 2–24 GB VRAM
- Min VRAM
- 2 GB
- GPU class
- Entry GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
OpenAI Whisper
The reference open-source speech-to-text model.
Open Source 2–10 GB VRAM
- Min VRAM
- 2 GB
- GPU class
- Entry GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
faster-whisper
Whisper, 4× faster, same accuracy. CTranslate2 backend.
Open Source 2–6 GB VRAM
- Min VRAM
- 2 GB
- GPU class
- Entry GPU
- Quant
- INT8
Actually FreeNo SignupOpen SourceWatermark-Free
FaceFusion
The most active open face-swap toolkit.
Open Source 2–8 GB VRAM
- Min VRAM
- 2 GB
- GPU class
- Entry GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Real-ESRGAN
The default OSS upscaler, still.
Open Source 2–6 GB VRAM
- Min VRAM
- 2 GB
- GPU class
- Entry GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free