All evidence cases
Terminal-Bench 2.1terminalPython

Implement AFAB pipeline parallelism for LLaMA

Implement one LLaMA training step that balances layers across ranks and executes all microbatch forwards before all backwards with correct distributed tensors.

torch-pipeline-parallelism

Normalized profile

What this task asks for

Implement one LLaMA training step that balances layers across ranks and executes all microbatch forwards before all backwards with correct distributed tensors. Export train_step_pipeline_afab with the specified signature. Partition layers roughly evenly for world sizes one and two. Perform every forward before any backward. Communicate hidden states and gradients with the specified shapes. Compute and microbatch-scale cross-entropy on the last rank without hooks.

  • feature implementation
  • performance optimization
  • test and validation
  • distributed training
  • model layers
  • tensor communication
  • PyTorch
  • torch.distributed
  • LLaMA
  • pipeline parallelism
  • /app/pipeline_parallel.py
  • concurrency, cancellation, or distributed-state coordination
  • changes span multiple components or interfaces

Published evidence

Model × harness outcomes

Counts are tied to these exact configurations. Harnesses are not treated as equivalent across benchmarks.

Models
13
Configurations
20
Trials
100

Showing 20 of 20 configurations

ModelHarnessEffortResultPasses
claude-fable-5anthropic · 2026-06-07claude-code2.1.167xhighAll passed5/5
claude-fable-5anthropic · 2026-06-05terminus-22.0.0highAll passed5/5
claude-opus-4-7anthropic · 2026-05-01claude-code2.1.123maxMixed1 errors2/5
claude-opus-4-7anthropic · 2026-05-01terminus-22.0.0maxNo passes4 errors0/5
claude-opus-4-8anthropic · 2026-07-09claude-code2.1.205highAll passed5/5
claude-sonnet-5anthropic · 2026-07-09claude-code2.1.205highMixed3/5
gemini-3-pro-previewgemini · 2026-05-01gemini-cli0.40.0highNo passes1 errors0/5
gemini-3-pro-previewgemini · 2026-05-01terminus-22.0.0highMixed3/5
gemini-3.1-pro-previewgemini · 2026-05-05gemini-cli0.40.0highNo passes0/5
gemini-3.1-pro-previewgemini · 2026-05-05terminus-22.0.0highMixed1 errors1/5
glm-5.1zai · 2026-05-01claude-code2.1.123maxNo passes5 errors0/5
gpt-5.5openai · 2026-05-01codex0.125.0xhighAll passed5/5
gpt-5.5openai · 2026-05-01terminus-22.0.0xhighMixed4 errors1/5
gpt-5.6-lunaopenai · 2026-07-10codex0.144.0maxMixed1 disqualified2/5
gpt-5.6-lunaopenai · 2026-07-11codex0.144.1maxAll passed5/5
gpt-5.6-solopenai · 2026-07-10codex0.144.0maxNo passes5 disqualified0/5
gpt-5.6-terraopenai · 2026-07-10codex0.144.0maxMixed3 disqualified1/5
gpt-5.6-terraopenai · 2026-07-11codex0.144.1maxNo passes0/5
grok-4.5cursor · 2026-07-09cursor-cli2026.07.08-0c04a8adefaultNo passes4 disqualified0/5
muse-spark-1.1openai · 2026-07-09mini-swe-agent2.4.5xhighNo passes0/5