Skip to content

Navigation Menu

Appearance settings

View all features
- 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
- 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
- COMMUNITY
  GitHub SponsorsFund open source developers
- PROGRAMS
  Security Lab
  Maintainer Community
  Accelerator
  Archive Program
- REPOSITORIES
  Topics
  Trending
  Collections
- ENTERPRISE SOLUTIONS
  Enterprise platformAI-powered developer platform
- AVAILABLE ADD-ONS
  GitHub Advanced SecurityEnterprise-grade security features
  Copilot for BusinessEnterprise-grade AI features
  Premium SupportEnterprise-grade 24/7 support
Pricing

Search code, repositories, users, issues, pull requests...

Search

Clear

Search syntax tips

Provide feedback

We read every piece of feedback, and take your input very seriously.

Include my email address so I can be contacted

Saved searches

Use saved searches to filter your results more quickly

Name

Query

To see all available qualifiers, see our documentation.

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

PrimeIntellect-ai / verifiers Public

Notifications You must be signed in to change notification settings
Fork 504
Star 3.9k

Code
Issues 60
Pull requests 96
Discussions
Actions
Projects
Security
Insights

Additional navigation options

Code
Issues
Pull requests
Discussions
Actions
Projects
Security
Insights

Comments

Close side panel

Support for token-in vLLM endpoint#626

Merged

mikasenghaas merged 84 commits intomainPrimeIntellect-ai/verifiers:mainfrom

tok-in-outPrimeIntellect-ai/verifiers:tok-in-outCopy head branch name to clipboard

Dec 16, 2025

Conversation Commits84 (84)Checks Files changed

Merged

Support for token-in vLLM endpoint#626
mikasenghaas merged 84 commits intomainfrom
tok-in-out

Commits

Commits on Dec 16, 2025

skeleton of token in multi-turn conversation
mikasenghaas
committed
fix url parsing
mikasenghaas
committed
allow configuring in eval entrypoint
mikasenghaas
committed
raise for status
mikasenghaas
committed
set sampling args and print warning
mikasenghaas
committed
correctly process tools
mikasenghaas
committed
fix eval cli test
mikasenghaas
committed
use setter
mikasenghaas
committed
remove prints
mikasenghaas
committed
create a oai client copy with diff base url and use post from there to use retires/ timeouts directly
mikasenghaas
committed
also use oai post routes for tokenize
mikasenghaas
committed
use incremental tokenization trick again
mikasenghaas
committed
suffix trick
mikasenghaas
committed
only call env_response once
mikasenghaas
committed
only check suffix if and eom token present
mikasenghaas
committed
parallelize tokenization calls
mikasenghaas
committed
avoid redundant concat
mikasenghaas
committed
fix typo
mikasenghaas
committed
move use token prompts into state
mikasenghaas
committed
add use token prompts to metadata
mikasenghaas
committed
removed docstring
mikasenghaas
committed
merge logic into get_model_response to avoid redundant code
mikasenghaas
committed
abstract tokenize_vllm and do not use token prompts on initial turn
mikasenghaas
committed
allow local tokenization to save http overhead
mikasenghaas
committed
rename to utp
mikasenghaas
committed
better best case complexity for find_lst_index
mikasenghaas
committed
configure exact tokenization
mikasenghaas
committed
fix caches + one run local tokenizer in process pool
mikasenghaas
committed
add debug logs
mikasenghaas
committed
generate client
mikasenghaas
committed
reverse conditional
mikasenghaas
committed
use thread pool
mikasenghaas
committed
fix caching bug
mikasenghaas
committed
cleanup
mikasenghaas
committed
more workers in threadpool
mikasenghaas
committed
do exact tokenization
mikasenghaas
committed
bring back setter
mikasenghaas
committed
match signature
mikasenghaas
committed
setup state from class attr
mikasenghaas
committed
also set tokenize method
mikasenghaas
committed
avoid mutation
mikasenghaas
committed
larger default thread pool
mikasenghaas
committed
make exact tokenization configurable
mikasenghaas
committed
fix passing exact tokenization
mikasenghaas
committed
use client directly on /v1/chat/completions/tokens route
mikasenghaas
committed
add timing [revert later]
mikasenghaas
committed
Revert "add timing [revert later]"
mikasenghaas
committed
add exact tokenization in rollout
mikasenghaas
committed
make tokens prompt arg none by default
mikasenghaas
committed
compute + cache suffix ids in non-exact tokenization and make it the default
mikasenghaas
committed
fix tests
mikasenghaas
committed
fix ty
mikasenghaas
committed
fix caching edge case with truncated turns
mikasenghaas
committed
correctly support completions tokenization req
mikasenghaas
committed
shorter warning
mikasenghaas
committed
fix ty
mikasenghaas
committed
only support setting token prompt args via class attrs/ setters
mikasenghaas
committed
do not tokenize with tools in non exact mode
mikasenghaas
committed
fix overlap
mikasenghaas
committed
fix adding suffix
mikasenghaas
committed
remove tokenize_method
mikasenghaas
committed
move tokenize method
mikasenghaas
committed
deprecate exact tokenization
mikasenghaas
committed
move building prompt_ids into get_model_respsonse
mikasenghaas
committed
using generic extra env kwargs setter
mikasenghaas
committed
fix tests
mikasenghaas
committed
revert env group changes
mikasenghaas
committed
revert moving line
mikasenghaas
committed
remove unused logger
mikasenghaas
committed
revert changing state
mikasenghaas
committed
abstract overlong prompt handling into decorator
mikasenghaas
committed
rename to get_model_response_with_messages
mikasenghaas
committed
fix tokens_client caching
mikasenghaas
committed
fix typo
mikasenghaas
committed
more typo
mikasenghaas
committed
more accurate comment
mikasenghaas
committed
allow setter
mikasenghaas
committed
fix url parsing edge case
mikasenghaas
committed
support generic setters
mikasenghaas
committed
move method
mikasenghaas
committed
more readable find_last_index
mikasenghaas
committed
fix
mikasenghaas
committed
remove the is not recommended for general use
mikasenghaas
committed
fix find last index again
mikasenghaas
committed

Footer

© 2026 GitHub, Inc.

Footer navigation

Terms
Privacy
Security
Status
Community
Docs
Contact

You can’t perform that action at this time.