Skip to content
Speculators Docs
Evaluating Model Performance
Initializing search
GitHub
Home
User Guide
Developer Guide
API Reference
CLI Reference
Community
Speculators Docs
GitHub
Home
User Guide
User Guide
Getting Started
Features
Algorithms
Algorithms
Eagle3
Dflash
Decision Guide
Tutorials
Tutorials
Train Eagle3 Model Online
Train Eagle3 Model Offline
Train Dflash Model Online
Response Regeneration
Evaluating Model Performance
Serve in vLLM
Developer Guide
Developer Guide
Contributing Guide
Code of Conduct
Add New Spec Decode Algorithm
Design Documents
Branding Guidelines
API Reference
API Reference
speculators
speculators
config
model
version
convert
convert
entrypoints
eagle
eagle
eagle3_converter
eagle3_legacy_model
eagle_converter
eagle_legacy_model
utils
data_generation
data_generation
configs
logging_utils
preprocessing
vllm_client
models
models
attention
base_components
utils
dflash
dflash
attention
config
core
metrics
model_definitions
utils
eagle3
eagle3
attention
config
core
data
model_definitions
proposals
proposals
base
greedy
train
train
checkpointer
data
distributed_batch_sampler
logger
noise_transforms
trainer
utils
vocab_mapping
utils
utils
loading
pydantic_utils
registry
util
CLI Reference
CLI Reference
prepare_data.py
data_generation_offline.py
launch_vllm.py
train.py
response_regeneration
Community
Community
Contact Us
Events & Meetups
Slack
Blog
Home
User Guide
Tutorials
Evaluating Model Performance
Coming soon
Back to top