Run FastGen inference
Apply fastgen-runtime. The
inference README owns the scripts, arguments, modes, checkpoint
selection, guidance and FID evaluation; the ComfyUI nodes provide
the interactive route. Read weights and sampling
for checkpoint roles and reproducible comparisons.
For preparation-only requests, provide the command and pending checks without launching.
- Resolve. Config, teacher versus student (or both), weight source, prompts or class
labels, conditioning media, seed, sampling settings and destination. Verify that the
checkpoint matches the network, method and branch, and never present an initialised
student as distilled (checkpoint selection).
- Entry point. Pick the script or node that supports the task and preserves every
output modality; inspect its
--help and module docstring. Teacher steps come from
--sample_steps, student steps from the scheduler's model.student_grid_cfg.steps
(key arguments); methods can add constraints, so
verify the effective number of network evaluations.
- Inputs. Check list lengths and pairing: class labels within the configured count,
source images for editing, image, video or camera inputs for video modes, and the
network's negative conditioning (guidance). Short
lists repeat their last entry, so count the effective inputs.
- Run. Use explicit output paths and the requested checkpoint within the resource
budget. For parity or acceptance evidence follow
fastgen-validate.
- Inspect. Open the media and native outputs; check dimensions, temporal alignment,
conditioning and sample count, and record the loaded weight source. Local files do not
prove tracker uploads.
State whether inference was prepared or executed; report artifact paths, sampling branch
and checkpoint, seed, grid, guidance and input settings. For memory failures inspect
resident networks and offload options before changing the workload; for bad samples
compare preprocessing and schedule against the reference before blaming the weights
(inference ignores a checkpoint).
1---2name: fastgen-inference3description: >-4 Generate and inspect NVIDIA FastGen teacher or distilled-student outputs with the inference scripts or ComfyUI nodes: config, checkpoint, prompts, conditioning media, seeds and sampling settings. Use it to prepare or execute sampling; use fastgen-validate to assess existing parity or quality evidence.5license: Apache-2.06metadata:7 source: https://skillmd.com/skills/nvidia/fastgen-inference8 author: "Julius Berner <jberner@nvidia.com>"9 domain: ml10---1112# Run FastGen inference1314Apply [fastgen-runtime](https://github.com/NVlabs/FastGen/blob/main/skills/fastgen-runtime/SKILL.md). The15[inference README](https://github.com/NVlabs/FastGen/blob/main/scripts/README.md) owns the scripts, arguments, modes, checkpoint16selection, guidance and FID evaluation; the [ComfyUI nodes](https://github.com/NVlabs/FastGen/blob/main/comfyui/README.md) provide17the interactive route. Read [weights and sampling](references/weights-and-sampling.md)18for checkpoint roles and reproducible comparisons.19For preparation-only requests, provide the command and pending checks without launching.20211. **Resolve.** Config, teacher versus student (or both), weight source, prompts or class22 labels, conditioning media, seed, sampling settings and destination. Verify that the23 checkpoint matches the network, method and branch, and never present an initialised24 student as distilled ([checkpoint selection](https://github.com/NVlabs/FastGen/blob/main/scripts/README.md#checkpoint-selection)).252. **Entry point.** Pick the script or node that supports the task and preserves every26 output modality; inspect its `--help` and module docstring. Teacher steps come from27 `--sample_steps`, student steps from the scheduler's `model.student_grid_cfg.steps`28 ([key arguments](https://github.com/NVlabs/FastGen/blob/main/scripts/README.md#key-arguments)); methods can add constraints, so29 verify the effective number of network evaluations.303. **Inputs.** Check list lengths and pairing: class labels within the configured count,31 source images for editing, image, video or camera inputs for video modes, and the32 network's negative conditioning ([guidance](https://github.com/NVlabs/FastGen/blob/main/scripts/README.md#guidance)). Short33 lists repeat their last entry, so count the effective inputs.344. **Run.** Use explicit output paths and the requested checkpoint within the resource35 budget. For parity or acceptance evidence follow36 [fastgen-validate](https://github.com/NVlabs/FastGen/blob/main/skills/fastgen-validate/SKILL.md).375. **Inspect.** Open the media and native outputs; check dimensions, temporal alignment,38 conditioning and sample count, and record the loaded weight source. Local files do not39 prove tracker uploads.4041State whether inference was prepared or executed; report artifact paths, sampling branch42and checkpoint, seed, grid, guidance and input settings. For memory failures inspect43resident networks and offload options before changing the workload; for bad samples44compare preprocessing and schedule against the reference before blaming the weights45([inference ignores a checkpoint](https://github.com/NVlabs/FastGen/blob/main/fern/docs/troubleshooting.mdx#inference-appears-to-ignore-a-checkpoint)).