Official Inference Code of EAGLE-4
chore: release 0.1.1
Seyfal Sultanovdcea6c4b32ee
diff --git a/CHANGELOG.md b/CHANGELOG.mdindex 166fb5a..1301a36 100644--- a/CHANGELOG.md+++ b/CHANGELOG.md@@ -1,5 +1,47 @@# Changelog+## 0.1.1 (2026-10-21)+++### ⚠ BREAKING CHANGES++* **core:** streamlined configuration: `"mode": "eval"`, `"ragged"`, `cost_model.ragged_*` and `draft_model.moe_gate_acc` are removed, and a config that sets them is refused at load with a pointer to the [migration guide](docs/migration.md).++### Features++* **sglang:** serve with only the draft path by default, on SGLang's default overlap scheduler+* **sglang:** install the SGLang CUDA engine `0.5.23.dev252+eagle.5`+* **tokenspeed:** install the TokenSpeed engine `0.1.1+eagle.5` (experimental)+* **tokenspeed:** negotiate the plugin contract by capabilities+* **vllm:** answer vLLM's suspend_drafting and verify_budget hooks+* **core:** fp32 arbiter for greedy-parity divergences above the near-tie gap; a candidate that is the fp32 argmax passes+* **pkg:** install ROCm from public indexes+++### Bug fixes++* **sglang:** capture first-use graphs without flushing the allocator, so the first long-context requests start faster+* **sglang:** compile first-use kernels and the draft softmax at startup, and keep slow rounds out of live costs+* **core:** bound the effect of one slow round on a live-cost cell, and pool seed minima across context intervals+* **sglang:** size 16-token chain rounds with the chain-depth rule; narrower chains keep the tree rule+* **sglang:** record sampled proposals for recursive tree verification, and require complete sampled-distribution evidence+* **sglang:** accept the engine hooks' msgspec accept types and the relocated kv-indices kernel+* **kernels:** freeze the draft window at the last committed position+* **vllm:** measure served tree acceptance, apply measured tree rounds in serve mode, and choose the tree's narrowing batch size from round-weighted gains+* **vllm:** accept chain widths and tree rounds only when they were measured on the same setup+* **vllm:** serve without vLLM's varlen verify-graph flag+* **vllm:** show plugin messages in the server log through vLLM's logger+* **tokenspeed:** one serve default and one complete-schedule rule+* **pkg:** make sampling_check accept a no-spec server as its own reference, and load tools/sampling_check.py without requests+* **core:** import the parity HTTP client only when the fp32 arbiter reads from a parity run+* **pkg:** lock the published eagle-harness 0.1.0 wheel and the complete runtime for ROCm+++### Performance++* **sglang:** sample temperature-1 trees from recorded proposals in one fused Philox draw+* **sglang:** use the engine's prefill CUDA graph in the release recipes+## 0.1.0 (2026-10-20)- Serve EAGLE-4 drafts with SGLang on NVIDIA B300 and AMD MI355X.diff --git a/eagle_harness/__init__.py b/eagle_harness/__init__.pyindex 801ff71..6dcd9a8 100644--- a/eagle_harness/__init__.py+++ b/eagle_harness/__init__.py@@ -10,4 +10,4 @@ Layout:sglang/ sglang plugin: algorithm (hooks), workers, draft steps, tree attention, draft model"""-__version__ = "0.1.0" # x-release-please-version+__version__ = "0.1.1" # x-release-please-versiondiff --git a/pyproject.toml b/pyproject.tomlindex e7d8854..cd1cc56 100644--- a/pyproject.toml+++ b/pyproject.toml@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"[project]name = "eagle-harness"-version = "0.1.0"+version = "0.1.1"description = "EAGLE-Harness: cost-model-driven dynamic-tree speculative decoding (out-of-tree sglang, vLLM and TokenSpeed plugin)"readme = "README.md"license = "LicenseRef-Eagle-4-Source-Available"