Skip to main content
⌘K
Drop a repoSign up
NEWvicrodh/qbzjust droppedNEWkowalski-ptr/compass-before-compactjust droppedNEWlastmile-ai/mcp-agentjust droppedREPOstablyai/orca+380 stars 24hREPOdiegosouzapw/OmniRoute+297 stars 24hREPOearendil-works/pi+262 stars 24hREPOpbakaus/impeccable+232 stars 24hREPOkoala73/worldmonitor+212 stars 24hNEWvicrodh/qbzjust droppedNEWkowalski-ptr/compass-before-compactjust droppedNEWlastmile-ai/mcp-agentjust droppedREPOstablyai/orca+380 stars 24hREPOdiegosouzapw/OmniRoute+297 stars 24hREPOearendil-works/pi+262 stars 24hREPOpbakaus/impeccable+232 stars 24hREPOkoala73/worldmonitor+212 stars 24h
THUDM/AgentBenchpublicPythongithub.com/THUDM/AgentBench
THUDM avatar

AgentBench

A Comprehensive Benchmark to Evaluate LLMs as Agents (ICLR'24)

AI AGENTSLLM INFRACHATGPTGPT 4LLMLLM AGENTTRENDING #2854
Comments0CompareSet alertExportRSSVisit repo

// OWNER · THUDM · REPO SNAPSHOT

github.com/THUDM

THUDM

A Comprehensive Benchmark to Evaluate LLMs as Agents (ICLR'24)

GitHubTHUDM
STARS3,603+18 · 7d
FORKS272+0 · 7d
CONTRIBUTORS0maintainer graph thin
OPEN ISSUES · PRS63· 0 
LICENSE—no license file
LANGUAGEPython 
FOUNDED—
LOCATION—
MEMBERS—
PUBLIC REPOS—
TOTAL STARS—
FOCUS—
FUNDING—

//STAR HISTORY · 12M

+115 all-time-window
Stars · cumulative3,603 · today
//WHY IT'S TRENDING· narrative synthesis · medium confidence

THUDM/AgentBench is sitting at #2854 on the trending leaderboard with a pulse of 2/100 with no cross-source channels firing yet — GitHub-stars-only signal so far.

The 7-day star delta is +18 (+0.5%) against a base of 3,603, so the headline number is category momentum more than a personal breakout. What's actually moving is the AI agent / LLM tooling stack.

Watch-outs: no tagged release on record (treat as pre-stable).

↻ refreshed 30m ago· next sync in 35m· sourced from 0 channels · 0 mentions / 24h

// RELATED REPOS · CROSS-SOURCE OVERLAP

5 detected · 30d window
#01
langchain-ai/langchain
agent-memoryai-agentsclaude-codeShared tag: agent-memory.
CO-MENTIONS · 30D (4 total)
4
★ 143k
30
last · 5h ago
#02
unslothai/unsloth
ai-agentsllm-infraShared tag: ai-agents.
CO-MENTIONS · 30D (13 total)
310
★ 68.9k
29
last · 2h ago
#03
ZhuLinsen/daily_stock_analysis
ai-agentsllm-infraShared tag: ai-agents.
CO-MENTIONS · 30D (4 total)
31
★ 59.1k
29
last · 12h ago
#04
AstrBotDevs/AstrBot
ai-agentsllm-inframcpShared tag: ai-agents.
CO-MENTIONS · 30D (9 total)
9
★ 38.2k
28
last · 4h ago
#05
HKUDS/Vibe-Trading
ai-agentsllm-inframcpShared tag: ai-agents.
CO-MENTIONS · 30D (21 total)
21522
★ 27.9k
28
last · 12h ago

▌ FAQ · answers from the data spine

What is THUDM/AgentBench?
A Comprehensive Benchmark to Evaluate LLMs as Agents (ICLR'24)
Who maintains AgentBench?
Maintained by THUDM (the owner) on GitHub.
What language is AgentBench written in?
Primarily Python.
When was AgentBench created? When was the last update?
Created July 28, 2023 · last commit February 8, 2026.
How do I install AgentBench?
Clone the repo:
git clone https://github.com/THUDM/AgentBench.git
Then follow the README in the cloned directory.
Is AgentBench actively maintained?
stalled — no commits for an extended period

//COMMENTS · 0

live thread · operators only

Sign in to join the discussion

Be the first to comment.
Created
Updated
Last commit
RepositoryTHUDM/AgentBench