Skip to content
Navigation Menu
Sign in
Appearance settings
Platform
AI CODE CREATION
GitHub Copilot
Write better code with AI
GitHub Copilot app
Direct agents from issue to merge
MCP Registry
Integrate external tools
DEVELOPER WORKFLOWS
Actions
Automate any workflow
Codespaces
Instant dev environments
Issues
Plan and track work
Code Review
Manage code changes
Code Quality
Enforce quality at merge
APPLICATION SECURITY
GitHub Advanced Security
Find and fix vulnerabilities
Code security
Secure your code as you build
Secret protection
Stop leaks before they start
EXPLORE
Why GitHub
Documentation
Blog
Changelog
Marketplace
View all features
Solutions
BY COMPANY SIZE
Enterprises
Small and medium teams
Startups
Nonprofits
BY USE CASE
App Modernization
DevSecOps
DevOps
CI/CD
View all use cases
BY INDUSTRY
Healthcare
Financial services
Manufacturing
Government
View all industries
View all solutions
Resources
EXPLORE BY TOPIC
AI
Software Development
DevOps
Security
View all topics
EXPLORE BY TYPE
Customer stories
Events & webinars
Ebooks & reports
Business insights
GitHub Skills
SUPPORT & SERVICES
Documentation
Customer support
Community forum
Trust center
Partners
View all resources
Open Source
COMMUNITY
GitHub Sponsors
Fund open source developers
PROGRAMS
Security Lab
Maintainer Community
GitHub Stars
Archive Program
REPOSITORIES
Topics
Trending
Collections
Enterprise
ENTERPRISE SOLUTIONS
Enterprise platform
AI-powered developer platform
AVAILABLE ADD-ONS
GitHub Advanced Security
Enterprise-grade security features
Copilot for Business
Enterprise-grade AI features
Premium Support
Enterprise-grade 24/7 support
Pricing
Search
/
Sign in
Sign up
Appearance settings
You signed in with another tab or window.
Reload
to refresh your session.
You signed out in another tab or window.
Reload
to refresh your session.
You switched accounts on another tab or window.
Reload
to refresh your session.
Dismiss alert
{{ message }}
NVIDIA
/
TensorRT-LLM
Public
Notifications
You must be signed in to change notification settings
Fork
2.8k
Star
14.7k
Code
Issues
606
Pull requests
890
Discussions
Actions
Projects
Security and quality
0
Insights
Additional navigation options
Code
Issues
Pull requests
Discussions
Actions
Projects
Security and quality
Insights
[None][feat] Add Qwen3-based DSpark drafter (DeepSpec dense checkpoints)
- #16813
#16813
Closed
chungen04
wants to merge 8 commits into
NVIDIA:main
NVIDIA/TensorRT-LLM:main
from
chungen04:dspark-qwen3
chungen04/TensorRT-LLM:dspark-qwen3
Copy head branch name to clipboard
Conversation
Commits
8
(8)
Checks
Files changed
Closed
[None][feat] Add Qwen3-based DSpark drafter (DeepSpec dense checkpoints)
#16813
chungen04
wants to merge 8 commits into
NVIDIA:main
NVIDIA/TensorRT-LLM:main
from
chungen04:dspark-qwen3
chungen04/TensorRT-LLM:dspark-qwen3
Copy head branch name to clipboard
Commits
Commits on Aug 5, 2026
[None][feat] Add Qwen3-based DSpark drafter (DeepSpec dense checkpoints)
Show description for 40289c8
chungen04
committed
40289c8
View commit details
Copy full SHA for 40289c8
Browse repository at this point
[None][refactor] blend Qwen3-based speculator support into existing framework
Show description for a23949d
chungen04
committed
a23949d
View commit details
Copy full SHA for a23949d
Browse repository at this point
Commits on Aug 6, 2026
trim comments
Show description for f301332
chungen04
committed
f301332
View commit details
Copy full SHA for f301332
Browse repository at this point
Commits on Aug 17, 2026
[None][fix] fail fast when DSpark mask_token_id can't be resolved from checkpoint or config
Show description for d820ad7
chungen04
committed
d820ad7
View commit details
Copy full SHA for d820ad7
Browse repository at this point
[None][fix] warn once when serving max_seq_len exceeds the DSpark drafter's trained RoPE range, and reject rope_scaling drafter configs
Show description for cc82644
chungen04
committed
cc82644
View commit details
Copy full SHA for cc82644
Browse repository at this point
[None][fix] gate DSpark's unprefixed config.json fallback on the checkpoint's Qwen3DSpark architecture
Show description for df30e8a
chungen04
committed
df30e8a
View commit details
Copy full SHA for df30e8a
Browse repository at this point
[None][test] add a single-GPU DSpark accuracy test for Qwen3-8B mirroring test_dflash
Show description for 2f43213
chungen04
committed
2f43213
View commit details
Copy full SHA for 2f43213
Browse repository at this point
[None][feat] flow the DSpark drafter's context window through DSparkDecodingConfig.ctx_window_size instead of an environment variable
Show description for ad5f580
chungen04
committed
ad5f580
View commit details
Copy full SHA for ad5f580
Browse repository at this point
You can’t perform that action at this time.