Instructions to use LocalAI-io/GEM-X-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use LocalAI-io/GEM-X-GGUF with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf LocalAI-io/GEM-X-GGUF:F32 # Run inference directly in the terminal: llama cli -hf LocalAI-io/GEM-X-GGUF:F32
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf LocalAI-io/GEM-X-GGUF:F32 # Run inference directly in the terminal: llama cli -hf LocalAI-io/GEM-X-GGUF:F32
Use pre-built binary
# Download pre-built binary from: # https://github.com/ggerganov/llama.cpp/releases # Start a local OpenAI-compatible server with a web UI: ./llama-server -hf LocalAI-io/GEM-X-GGUF:F32 # Run inference directly in the terminal: ./llama-cli -hf LocalAI-io/GEM-X-GGUF:F32
Build from source code
git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-server llama-cli # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama-server -hf LocalAI-io/GEM-X-GGUF:F32 # Run inference directly in the terminal: ./build/bin/llama-cli -hf LocalAI-io/GEM-X-GGUF:F32
Use Docker
docker model run hf.co/LocalAI-io/GEM-X-GGUF:F32
- LM Studio
- Jan
- Ollama
How to use LocalAI-io/GEM-X-GGUF with Ollama:
ollama run hf.co/LocalAI-io/GEM-X-GGUF:F32
- Unsloth Desktop
- Docker Model Runner
How to use LocalAI-io/GEM-X-GGUF with Docker Model Runner:
docker model run hf.co/LocalAI-io/GEM-X-GGUF:F32
- Lemonade
How to use LocalAI-io/GEM-X-GGUF with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull LocalAI-io/GEM-X-GGUF:F32
Run and chat with the model
lemonade run user.GEM-X-GGUF-F32
List all available models
lemonade list
- Atomic Chat
Download reference/sources.json from LocalAI-io/GEM-X-GGUF: direct link, hf CLI and curl.
- Browser
- Download file 4.73 kB
-
https://huggingface.co/LocalAI-io/GEM-X-GGUF/resolve/main/reference/sources.json
- Command line
-
hf download hf://LocalAI-io/GEM-X-GGUF/reference/sources.json
-
curl -L -o sources.json https://huggingface.co/LocalAI-io/GEM-X-GGUF/resolve/main/reference/sources.json
4.73 kB
| { | |
| "schema_version": 1, | |
| "source": { | |
| "repository": "https://github.com/NVlabs/GEM-X", | |
| "revision": "32992550dba114c62243fb55e361311972dce8f9", | |
| "license": "Apache-2.0", | |
| "sam3d_body_revision": "b5c765a0d89d789985e186d396315e7590887b94", | |
| "soma_revision": "e0f8ff0ecfa3edbbb6058b1e0f08822ee2f84ee5", | |
| "motion_stats_sha256": "dafe4ef6a62e824b0b325f54e42d508015785509bc478ef76e4208ae1f95ba7c" | |
| }, | |
| "official_demo": { | |
| "note": "NVIDIA publishes rendered GIFs, not the clean source video or numeric outputs. The keypoint GIF can be used as an annotated 72-frame smoke-test input, but it is not pixel-identical to the source used for the published renders.", | |
| "demo_2d_keypoints.gif": { | |
| "bytes": 5638315, | |
| "frames": 72, | |
| "fps": 10, | |
| "sha256": "b3c34367565c5ad770b77107d8cf39144d5a4a84ce758957341a2d4c87e25aef" | |
| }, | |
| "demo_global_mesh.gif": { | |
| "bytes": 1867225, | |
| "frames": 72, | |
| "fps": 10, | |
| "sha256": "7513f7b7a95113140ba77986c1141e26fe22d8612602b440eb4d94005594f766" | |
| }, | |
| "demo_incam_mesh.gif": { | |
| "bytes": 5611379, | |
| "frames": 72, | |
| "fps": 10, | |
| "sha256": "b18e198b5f26ad521e70e9c069fe73817115b81a6e0b8a285492e85f381668ee" | |
| } | |
| }, | |
| "paired_e2e": { | |
| "note": "Historical shared-input diagnostic: upstream reused native boxes/keypoints. Not independent end-to-end evidence. See docs/LIVE-OFFLINE-PARITY.md for the independent rerun.", | |
| "input_exact.mkv": { | |
| "bytes": 12805076, | |
| "frames": 72, | |
| "fps": 10, | |
| "sha256": "b7e2cff0f477d28267589fc04052aef311e362d8febf62527a0c97cf68c18d89" | |
| }, | |
| "hpe_results.pt": { | |
| "bytes": 1003741, | |
| "sha256": "02cd65e619b722443adaee7d73bcabf044b55ce5454b184cbab7868fdbbf98eb" | |
| }, | |
| "input_exact_1_incam.mp4": { | |
| "bytes": 797469, | |
| "sha256": "393d388153c2e54f80cb99725540faa7759326dc4bc7f75fdb7b800fb5db5324" | |
| }, | |
| "input_exact_2_global.mp4": { | |
| "bytes": 100921, | |
| "sha256": "08e58508920de81b985fadff75e627008a44a6a9a1df939df143333d00ed6d76" | |
| } | |
| }, | |
| "models": { | |
| "repository": "https://huggingface.co/nvidia/GEM-X", | |
| "revision": "5ccf5ca3746c3620aa4016114f069a5f6ae399cd", | |
| "gem_soma.ckpt": { | |
| "bytes": 541758499, | |
| "sha256": "4c1f85ca8c1e11e6588aead49fbc024bf660708def670043e0b537c101ee298e" | |
| }, | |
| "onnx/gem_denoiser.onnx": { | |
| "bytes": 1508810, | |
| "sha256": "20aab83c01bbd909a258ad0fa465458eda382c63dc80d56952b2ec952d6e192c" | |
| }, | |
| "onnx/gem_denoiser.onnx.data": { | |
| "bytes": 199114752, | |
| "sha256": "720c2400281b730574bf2a446c33ecd77bf494ac8161b081b09bb122b9fad2f8" | |
| }, | |
| "onnx/vitpose.onnx": { | |
| "bytes": 4279801, | |
| "sha256": "0982dbf4f1e8a48446a6fe35329711522b60210cce2a7499ca6ab93458c87f34" | |
| }, | |
| "onnx/vitpose.onnx.data": { | |
| "bytes": 3389390848, | |
| "sha256": "b20ba3077ba2341d76c16dde3e58d3b37c66c3b0ec8346901a8bac14319e20fe" | |
| }, | |
| "scale_comps.pth": { | |
| "bytes": 8816, | |
| "sha256": "4c4861384b176fbe3d3bd7480e87ff8161f19553bf6ec1b718bc9b6bbb559cd6" | |
| }, | |
| "scale_mean.pth": { | |
| "bytes": 1451, | |
| "sha256": "f696a0793217f27fcf23fa1fd63e8a34aa74ef64580078698fb9ea7322ce2891" | |
| }, | |
| "SOMA_neutral.npz": { | |
| "repository": "https://huggingface.co/nvidia/SOMA-X", | |
| "revision": "466879a", | |
| "bytes": 27488638, | |
| "sha256": "515f7d5bb74be4e370e9adf5e779760ec3581556374c0b33212a32d13ab3b53f" | |
| }, | |
| "onnx/gem_denoiser_no_imgfeat.onnx": { | |
| "bytes": 187936462, | |
| "sha256": "65c576968bc9548b5c485522930b7985097b70a5e0154e5cd01241d6c806415f" | |
| } | |
| }, | |
| "detector": { | |
| "model": "YOLOX-X HumanArt a39d44ed", | |
| "repository": "https://github.com/open-mmlab/mmpose", | |
| "url": "https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/onnx_sdk/yolox_x_8xb8-300e_humanart-a39d44ed.zip", | |
| "zip_bytes": 368200375, | |
| "zip_sha256": "1c6fccc5cd450eb563b1fe4eba13c9bdf9ea8263b6ae1cb3ada765aa4ebea2c7", | |
| "onnx_member": "20230928/yolox_onnx/yolox_x_8xb8-300e_humanart-a39d44ed/end2end.onnx", | |
| "onnx_bytes": 396283264, | |
| "onnx_sha256": "8e9ea96a176bd48501eaaa77216e49ee30794d2f8ba80c7b9862beca4ea972da" | |
| }, | |
| "live_example": { | |
| "url": "https://raw.githubusercontent.com/NVlabs/GR00T-WholeBodyControl/main/gear_sonic/examples/live_camera_teleop/webcam_stream.py", | |
| "captured": "2026-09-18", | |
| "sha256": "0342737f98302918e5fe10ca9536e08196277e2969845cf47608f60c0ae66fe2" | |
| }, | |
| "regression_fixtures": { | |
| "gem_contact_reference.bin": { | |
| "bytes": 168492, | |
| "sha256": "1127ad8ebf4686c66ff942a3fa8fb6c1977b50733dfec60b40dae65d444f9b23" | |
| }, | |
| "gem_live_reference.bin": { | |
| "bytes": 332388, | |
| "sha256": "a1127cd8a876c2b9c68158eb0383f707be95eab20d6afee6241ca901bd37aee4" | |
| } | |
| } | |
| } | |