メインコンテンツまでスキップ

Claims & Data Sources

Why TinyLoop Is Different From vLLM, TensorRT-LLM, And Similar Stacks

The point is not that those frameworks are weak. It is that they optimize a different default assumption.

QuestionTinyLoopGeneric runtime
Treats the shared loop block as a first-class runtime primitiveYesUsually no
Runtime-variable depth from one checkpointYesUsually not exposed cleanly
Single-checkpoint self-speculative decode (draft_loops vs verify_loops)YesNo, usually needs a second draft model or extra training
h-mode KV reconstruction from shared W_k / W_vYesNo, standard deep stacks do not have the same shared-weight property
Best choice for broad arbitrary-model serving todayNoYes
Best choice when the checkpoint family is already looped and weight-sharedYesOften not

Safe Claim vs Unsafe Claim

Safe claim:

  • TinyLoop is the better runtime when you want to preserve the actual structure of a looped transformer and monetize that structure in memory, cache design, and single-checkpoint decoding tricks.

Unsafe claim:

  • TinyLoop is universally faster than PyTorch, vLLM, TensorRT-LLM, or every other runtime on every workload.

This repo does not support the unsafe claim yet, and the page should not pretend otherwise.

Why Someone Should Choose TinyLoop

Choose TinyLoop when:

  • your model family is already weight-shared or looped
  • VRAM density is the first-order deployment constraint
  • you want depth to stay a runtime knob
  • you want loop-specific features such as h-mode KV storage or single-checkpoint self-speculative decode

Choose a generic runtime when:

  • you need wide checkpoint compatibility
  • you need mature batching and serving infrastructure immediately
  • the model is an ordinary deep stack with no loop-aware structure to preserve

Data Sources In This Repo

The numbers above are assembled from:

  • docs/framework/current-status.md
  • docs/framework/python-api/prefix-cache.md
  • docs/framework/python-api/generation.md
  • next.md
  • New (2026-04-19): tests/bench_marlin_gemm.cu, tests/test_marlin_parity.py, tests/test_marlin_asymmetric_parity.py, tests/test_marlin_asymm_correction_kernel.py, tests/bench_l2_residency.cu, plus the 4090 / 1B measurement bundle at paper_draft/4_19.md (§"Marlin INT4 end-to-end on RTX 4090" and §"Afternoon results — three big wins") and the raw logs under paper_draft/data/.