We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent 9fe49f1 commit 153bc70Copy full SHA for 153bc70
tests/integration/defs/accuracy/references/mmlu.yaml
@@ -143,7 +143,7 @@ Qwen3/Qwen3-8B:
143
- quant_algo: FP8_BLOCK_SCALES
144
accuracy: 76.12
145
- accuracy: 76.12
146
- - spec_dec_algo: EAGLE3
+ - spec_dec_algo: Eagle
147
148
Qwen3/Qwen3-30B-A3B:
149
tests/integration/defs/accuracy/test_llm_api_pytorch.py
@@ -1661,8 +1661,8 @@ def test_eagle3(self):
1661
1662
draft_len = 4
1663
spec_config = EagleDecodingConfig(max_draft_len=draft_len,
1664
- pytorch_weights_path=eagle_model_dir,
1665
- eagle3_one_model=False)
+ speculative_model_dir=eagle_model_dir
+ )
1666
1667
llm = LLM(model=target_model_dir,
1668
**pytorch_config,
0 commit comments