Skip to main content

← All model recipe tasks

elf_flow model family

Tasks

Supported architectures and task heads

The Hugging Face values below are copied from checkpoint metadata at the recorded revision. The TRTMC task contract comes from the exact E2E recipe. Architecture identity selects or describes a source graph; it does not imply that TRTMC reproduces every Hugging Face head with the same model_type. For example, an encoder recipe may consume the base model and intentionally return hidden states instead of the checkpoint's pretraining or classification logits.

HF model_typeHF architecture / pipeline classTRTMC task contractExact recipe profiles
ELF-B
Source: config.model
Metadata: ELF-B-de-en.yml at c8120fb80e52
diffusion_text_generation
elf-b-de-en-l0
elf-b-owt-l0
elf-b-xsum-l0

Declared recipes

Each row comes from a model-owned E2E manifest. It declares a test recipe; it is not a live pass receipt for every hardware target.

RecipeExact Hugging Face checkpointTaskBuild configurationDeclared E2E casesManifest
elf-b-de-en-l0embedded-language-flows/ELF-B-de-enfp16; single device
FP32 layers: 5, 6, 7, 8, 9, 10, 11
elf-b-de-en-l0tests/e2e/models/elf_flow/manifests/elf-b-de-en-l0.json
elf-b-owt-l0embedded-language-flows/ELF-B-owtfp32; single deviceelf-b-owt-l0tests/e2e/models/elf_flow/manifests/elf-b-owt-l0.json
elf-b-xsum-l0embedded-language-flows/ELF-B-xsumfp16; single device
FP32 layers: 5, 6, 7, 8, 9, 10, 11
elf-b-xsum-l0tests/e2e/models/elf_flow/manifests/elf-b-xsum-l0.json

Family-owned configuration

This family does not declare a family-owned --set namespace. Use the explicit CLI options shown below and the shared configuration namespaces documented in Configure Runtime Behavior.

Family-specific CLI contracts

Inputs and options below are filtered by the declared E2E task and the methods and configuration fields used by that family's native runtime implementation. The global CLI parser accepts a wider union of flags; flags absent here are not declared for this family.

trtmc run

Generate text with the family's diffusion-style text runtime.

Declared recipes: elf-b-de-en-l0, elf-b-owt-l0, elf-b-xsum-l0

trtmc run <bundle.bundle> --prompt "<text>" [text-diffusion options]
Supported input or optionRequirementRuntime behavior
--prompt <TEXT>Required unless initial latents are suppliedText conditioning input.
--max-new-tokens <N>OptionalLimit the number of generated tokens or audio frames.
--seed <N>OptionalSet the sampling or diffusion seed.
--num-steps <N>OptionalSet the number of diffusion or flow-matching steps.
--guidance-scale <F>OptionalSet classifier-free guidance strength.
--cfg-scale <F>OptionalSet the conditional classifier-free guidance scale.
--sde-gamma <F>OptionalSet the SDE or flow-matching gamma override.
--initial-latents-raw <PATH>OptionalLoad packed float32 initial latents.
--condition-latents-raw <PATH>OptionalLoad packed float32 conditioning latents.
--condition-mask-raw <PATH>OptionalLoad a float32 conditioning mask.
--sampling-steps-raw <PATH>OptionalLoad an explicit float32 sampling schedule.
--sde-noise-raw <PATH>OptionalLoad precomputed float32 SDE noise.
--output <PATH>OptionalWrite generated samples as JSON Lines.
Code basis

Runtime provider: elf_flow

  • include/trtmc/pipeline.h
  • src/cli/main.cpp
  • src/runtime/models/elf_flow/pipeline.cpp
  • tests/e2e/models/elf_flow/manifests/elf-b-de-en-l0.json
  • tests/e2e/models/elf_flow/manifests/elf-b-owt-l0.json
  • tests/e2e/models/elf_flow/manifests/elf-b-xsum-l0.json

Bundle lifecycle commands

These commands apply to every declared recipe in this family; they do not add task inputs or model capabilities.

trtmc build

Build one exact checkpoint into a TensorRT-Model-Connect bundle.

trtmc build <hf-id> --output <bundle.bundle>

trtmc inspect

Inspect bundle metadata, runtime identity, and packaged sections.

trtmc inspect <bundle.bundle>

See the CLI Reference for all options and limitations.