Model CoverageLarge Language ModelsByteDance SeedSeed-Coder-8B-Instruct

Seed-Coder-8B-Instruct

View as Markdown

Seed-Coder and Seed-OSS are open-weight models from ByteDance. Both use the Qwen2ForCausalLM architecture under the hood.

Set up NeMo AutoModel with the latest container or follow the installation instructions.

Fine-Tune Seed-Coder-8B-Instruct

From the repository root, run:

uv run automodel --nproc-per-node=8 examples/llm_finetune/seed/seed_coder_8b_instruct_squad.yaml

Choose a Workflow

GoalStart Here
Supervised fine-tuning (SFT) - Seed-Coder 8B on SQuADUse seed_coder_8b_instruct_squad.yaml.
Low-rank adaptation (LoRA) - Seed-Coder 8B on SQuADUse seed_coder_8b_instruct_squad_peft.yaml.

Model Reference

Model Architecture

PropertyValue
TaskText Generation
ArchitectureQwen2ForCausalLM
Parameters8B
Hugging Face OrganizationByteDance-Seed
  • Qwen2ForCausalLM (reuses Qwen2 architecture)

Available Models

ModelHF ID
Seed-Coder 8B InstructByteDance-Seed/Seed-Coder-8B-Instruct