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 }}
SemiAnalysisAI
/
InferenceX
Public
Notifications
You must be signed in to change notification settings
Fork
303
Star
1.7k
Code
Issues
103
Pull requests
179
Discussions
Actions
Projects
Security and quality
0
Insights
Additional navigation options
Code
Issues
Pull requests
Discussions
Actions
Projects
Security and quality
Insights
[AgentX] DeepSeek-V4-Pro-0813 llm-d on GB200
- #2719
#2719
Open
ilmarkov
wants to merge 58 commits into
main
SemiAnalysisAI/InferenceX:main
from
imarkov/dsv4-pro-agentic
SemiAnalysisAI/InferenceX:imarkov/dsv4-pro-agentic
Copy head branch name to clipboard
Conversation
Commits
58
(58)
Checks
Files changed
Open
[AgentX] DeepSeek-V4-Pro-0813 llm-d on GB200
#2719
ilmarkov
wants to merge 58 commits into
main
SemiAnalysisAI/InferenceX:main
from
imarkov/dsv4-pro-agentic
SemiAnalysisAI/InferenceX:imarkov/dsv4-pro-agentic
Copy head branch name to clipboard
Commits
Commits on Aug 24, 2026
Add DeepSeek-V4-Pro FP4 B200 llmd-vllm P/D disagg (agentX 1P-DEP8/1D-DEP8)
Show description for 8cf10ab
ilmarkov
and
cursoragent
committed
8cf10ab
View commit details
Copy full SHA for 8cf10ab
Browse repository at this point
Commits on Aug 26, 2026
Trim disagg conc-list to 192 and script the v0.10.0 EPP binary pull
Show description for 83c3060
ilmarkov
and
cursoragent
committed
83c3060
View commit details
Copy full SHA for 83c3060
Browse repository at this point
Add DeepSeek-V4-Pro FP4 B200 llmd-vllm aggregated (TP8/DEP8) benchmark
Show description for fe7e680
ilmarkov
and
cursoragent
committed
fe7e680
View commit details
Copy full SHA for fe7e680
Browse repository at this point
Commits on Aug 31, 2026
Update configs
ilmarkov
committed
79f3e12
View commit details
Copy full SHA for 79f3e12
Browse repository at this point
Merge upstream/main into imarkov/dsv4-pro-agentic
ilmarkov
committed
839c220
View commit details
Copy full SHA for 839c220
Browse repository at this point
Cleanup comments
ilmarkov
committed
8136823
View commit details
Copy full SHA for 8136823
Browse repository at this point
Address review comments
ilmarkov
committed
af3818d
View commit details
Copy full SHA for af3818d
Browse repository at this point
Update DSV4-Pro model to 0813 version
ilmarkov
committed
e648883
View commit details
Copy full SHA for e648883
Browse repository at this point
Commits on Sep 2, 2026
fix: run llm-d DSpark with AgentX golden acceptance and explicit metadata
cquil11
committed
a4edd45
View commit details
Copy full SHA for a4edd45
Browse repository at this point
merge: resolve main conflicts while preserving DSpark AgentX fixes
cquil11
committed
a437eb1
View commit details
Copy full SHA for a437eb1
Browse repository at this point
fix: accept cluster runner labels in llm-d container selection
cquil11
committed
e3d6d0c
View commit details
Copy full SHA for e3d6d0c
Browse repository at this point
chore: remove standalone llm-d recipe and redundant config tests
cquil11
committed
8d00f59
View commit details
Copy full SHA for 8d00f59
Browse repository at this point
docs: consolidate DSpark performance changelog
cquil11
committed
0391477
View commit details
Copy full SHA for 0391477
Browse repository at this point
fix: expose and verify llm-d worker metrics for AIPerf
cquil11
committed
a65c1e3
View commit details
Copy full SHA for a65c1e3
Browse repository at this point
fix(llm-d): use the B200 Nscale shared home for cached artifacts
cquil11
committed
f6cfad9
View commit details
Copy full SHA for f6cfad9
Browse repository at this point
fix(llm-d): match AgentX router metadata to the bundled version
cquil11
committed
1758eac
View commit details
Copy full SHA for 1758eac
Browse repository at this point
fix(llm-d): move AgentX to GB200 and simplify launch paths
cquil11
committed
4e5f6c6
View commit details
Copy full SHA for 4e5f6c6
Browse repository at this point
Adapt optimized configs and map to agentic gb200
ilmarkov
committed
775e5e9
View commit details
Copy full SHA for 775e5e9
Browse repository at this point
merge: resolve conflicts with upstream/imarkov/dsv4-pro-agentic
Show description for 07c6cfd
ilmarkov
and
claude
committed
07c6cfd
View commit details
Copy full SHA for 07c6cfd
Browse repository at this point
Fix recipe
ilmarkov
committed
95586e0
View commit details
Copy full SHA for 95586e0
Browse repository at this point
Fix spec decoding
ilmarkov
committed
6365015
View commit details
Copy full SHA for 6365015
Browse repository at this point
Address cursor's comment
ilmarkov
committed
11b8af5
View commit details
Copy full SHA for 11b8af5
Browse repository at this point
Commits on Sep 3, 2026
fix: keep GB200 import setup explicit and correct DSpark changelog
cquil11
committed
81060c2
View commit details
Copy full SHA for 81060c2
Browse repository at this point
merge: sync main while preserving the GB200 AgentX changelog entry
cquil11
committed
e8e1594
View commit details
Copy full SHA for e8e1594
Browse repository at this point
fix: preserve fixed DSpark draft budgets for golden AL
cquil11
committed
8493a29
View commit details
Copy full SHA for 8493a29
Browse repository at this point
fix: use GB200 InfiniBand adapters for Mooncake transfers
cquil11
committed
6394609
View commit details
Copy full SHA for 6394609
Browse repository at this point
fix: provide DEP8 router token-load dependencies
cquil11
committed
73476be
View commit details
Copy full SHA for 73476be
Browse repository at this point
fix: start a shared Mooncake master for llm-d jobs
cquil11
committed
3ba226b
View commit details
Copy full SHA for 3ba226b
Browse repository at this point
fix: reserve GPU memory for long-context llm-d replay
cquil11
committed
bc55cca
View commit details
Copy full SHA for bc55cca
Browse repository at this point
fix: cover TP8 and DEP8 long-context allocation peaks
cquil11
committed
9c1c0c4
View commit details
Copy full SHA for 9c1c0c4
Browse repository at this point
fix: prevent duplicate llm-d frontend metric scrapes
cquil11
committed
7b65775
View commit details
Copy full SHA for 7b65775
Browse repository at this point
fix: retain P/D KV blocks across long decode stalls
cquil11
committed
4a6231e
View commit details
Copy full SHA for 4a6231e
Browse repository at this point
fix: validate llm-d metrics targets and retain endpoint roles
cquil11
committed
9411ed9
View commit details
Copy full SHA for 9411ed9
Browse repository at this point
fix: copy configs for p1d1
ilmarkov
committed
1a31973
View commit details
Copy full SHA for 1a31973
Browse repository at this point
Add concurrency
ilmarkov
committed
70c8a39
View commit details
Copy full SHA for 70c8a39
Browse repository at this point
Commits on Sep 4, 2026
Adjust decode utilization
ilmarkov
committed
02ed639
View commit details
Copy full SHA for 02ed639
Browse repository at this point
fix
ilmarkov
committed
8080915
View commit details
Copy full SHA for 8080915
Browse repository at this point
fix prefill
ilmarkov
committed
da74471
View commit details
Copy full SHA for da74471
Browse repository at this point
Disable AV fin sake of golden acceptence
ilmarkov
committed
109f597
View commit details
Copy full SHA for 109f597
Browse repository at this point
Merge upstream/main into imarkov/dsv4-pro-agentic
Show description for 51255cb
ilmarkov
and
claude
committed
51255cb
View commit details
Copy full SHA for 51255cb
Browse repository at this point
Commits on Sep 14, 2026
Fix disagg prefill OOM and add SimpleCPUOffloadConnector to mooncake disagg
Show description for 4366bc2
ilmarkov
and
claude
committed
4366bc2
View commit details
Copy full SHA for 4366bc2
Browse repository at this point
Merge upstream/main into imarkov/dsv4-pro-agentic
Show description for 3e1b813
ilmarkov
and
claude
committed
3e1b813
View commit details
Copy full SHA for 3e1b813
Browse repository at this point
Move dsv4-gb200 changelog entry to end of file
Show description for 5be3fcf
ilmarkov
and
claude
committed
5be3fcf
View commit details
Copy full SHA for 5be3fcf
Browse repository at this point
Fix mock curl in test_llmd_agentic_adapter to handle --write-out preflight
Show description for b089e16
ilmarkov
and
claude
committed
b089e16
View commit details
Copy full SHA for b089e16
Browse repository at this point
Fix test_llmd_agentic_adapter endpoint fixture broken by 9411ed99b
Show description for 884b32b
ilmarkov
and
claude
committed
884b32b
View commit details
Copy full SHA for 884b32b
Browse repository at this point
Merge upstream/main into imarkov/dsv4-pro-agentic
Show description for 36c6f4d
ilmarkov
and
claude
committed
36c6f4d
View commit details
Copy full SHA for 36c6f4d
Browse repository at this point
Commits on Sep 15, 2026
Fix mooncake KV pollution and EPP routing for agentic p1d1 and DEP8
Show description for 23dae65
ilmarkov
and
claude
committed
23dae65
View commit details
Copy full SHA for 23dae65
Browse repository at this point
Merge remote-tracking branch 'upstream/main' into imarkov/dsv4-pro-agentic
ilmarkov
committed
893ba8c
View commit details
Copy full SHA for 893ba8c
Browse repository at this point
Correct after merge
ilmarkov
committed
d6a6d82
View commit details
Copy full SHA for d6a6d82
Browse repository at this point
Commits on Sep 16, 2026
fix(llm-d): use external LB and per-rank EPP endpoints for DEP8 AgentX
Show description for ad34e7f
ilmarkov
and
cursoragent
committed
ad34e7f
View commit details
Copy full SHA for ad34e7f
Browse repository at this point
Merge upstream/main into imarkov/dsv4-pro-agentic
Show description for fd8e4d5
ilmarkov
and
cursoragent
committed
fd8e4d5
View commit details
Copy full SHA for fd8e4d5
Browse repository at this point
Merge upstream/main into imarkov/dsv4-pro-agentic
ilmarkov
committed
a5eac8c
View commit details
Copy full SHA for a5eac8c
Browse repository at this point
Commits on Sep 17, 2026
fix(llm-d): restore GB200 disagg wrapper and re-deprecate 8k1k key
Show description for f88fe3c
ilmarkov
and
cursoragent
committed
f88fe3c
View commit details
Copy full SHA for f88fe3c
Browse repository at this point
Merge upstream/main into imarkov/dsv4-pro-agentic
Show description for 73cb898
ilmarkov
and
cursoragent
committed
73cb898
View commit details
Copy full SHA for 73cb898
Browse repository at this point
fix(llm-d): forward INFMAX_CONTAINER_WORKSPACE for agentic throughput
ilmarkov
committed
3d1ea2b
View commit details
Copy full SHA for 3d1ea2b
Browse repository at this point
debug(llm-d): enable VLLM_LOGGING_LEVEL=DEBUG for DSpark AgentX bring-up
Show description for 545e5e6
ilmarkov
and
cursoragent
committed
545e5e6
View commit details
Copy full SHA for 545e5e6
Browse repository at this point
fix(llm-d): shrink Mooncake DRAM budget and restore INFO logging
ilmarkov
committed
061901e
View commit details
Copy full SHA for 061901e
Browse repository at this point
Commits on Sep 18, 2026
Merge upstream/main into imarkov/dsv4-pro-agentic
Show description for 4742446
2 people
authored and
cquil11
committed
4742446
View commit details
Copy full SHA for 4742446
Browse repository at this point
You can’t perform that action at this time.
