Skip to content

Benchmark

AppWorld Test-Challenge Level 2: Task Goal Completion

Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.

Models and systems measured
26
Scores recorded
26
Sources
1
IQ weight
No weight published

Score distribution

33.5%
Mean
24%
Median
30.1%
Std dev
0.7%
10th percentile
84.7%
90th percentile
0%
Min
0%90.7%
The band covers the middle 80% of scores. The line marks the median.

Results

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

Model or systemScore
Claude Opus 4.6Button Agent (f072cd07-6ae0fbdf)· AppWorld Leaderboard data docs70%
claude-opus-4-7LARA MAS (23715209-d5c9ade3)· AppWorld Leaderboard data docs84.7%
DeepSeekCoderFullCodeRefl (0465049f-bff9fca2)· AppWorld Leaderboard data docs2%
DeepSeekCoderPlanExec (871bf9fd-4de66f57)· AppWorld Leaderboard data docs0%
DeepSeekCoderReAct (9ae299b8-25c255f5)· AppWorld Leaderboard data docs0%
GLM-5.3AtlasHarness (65df675b-ca5e4771)· AppWorld Leaderboard data docs86.7%
GLM-5.3-FlashIBM CUGA (a2db3f1e-2c28b1b2)· AppWorld Leaderboard data docs90.7%
GPT-4 TurboFullCodeRefl (571aa685-a8671c57)· AppWorld Leaderboard data docs9.3%
GPT-4 TurboIPFunCall (08d2ff3e-a0589430)· AppWorld Leaderboard data docs19.3%
GPT-4 TurboPlanExec (c722a232-74b9384b)· AppWorld Leaderboard data docs10.7%
GPT-4 TurboReAct (1f0055c0-123849bb)· AppWorld Leaderboard data docs16.7%
GPT-4.1IBM CUGA (8a02d49a-0df800d6)· AppWorld Leaderboard data docs58.7%
gpt-4.1-miniLARA MAS (695a4b55-1b68dde4)· AppWorld Leaderboard data docs36%
GPT-4oFullCodeRefl (d035381e-2d30c0e3)· AppWorld Leaderboard data docs18.7%
GPT-4oIPFunCall (3a889687-923696fe)· AppWorld Leaderboard data docs24%
GPT-4oPlanExec (ac5b742b-aa0243da)· AppWorld Leaderboard data docs17.3%
GPT-4oReAct (3414a75f-3df6a61f)· AppWorld Leaderboard data docs28.7%
GPT-4oReAct + 2 SetBSR Demos (1c12639a-57a0ac34)· AppWorld Leaderboard data docs34%
gpt-5.6-lunakecaipan capybara (1095b768-6513e0c6)· AppWorld Leaderboard data docs68.7%
gptoss-20bRooAGI_agentbuilder_memtune_rl (5acab4bb-badb978f)· AppWorld Leaderboard data docs11.3%
LLaMA3-70BFullCodeRefl (baf95530-cee120fc)· AppWorld Leaderboard data docs0.7%
LLaMA3-70BPlanExec (77f25849-862250c4)· AppWorld Leaderboard data docs0.7%
LLaMA3-70BReAct (4077db41-07bc5204)· AppWorld Leaderboard data docs3.3%
Qwen2.5-32BLOOP (e5e08b7e-d8e6474f)· AppWorld Leaderboard data docs40.7%
Qwen3-14BAlibaba Cloud ApsaraLab AgentRL (cabab065-0caee3c3)· AppWorld Leaderboard data docs62.7%
Qwen3.5-27BSynthAPI (3bd98253-d4fa6c8e)· AppWorld Leaderboard data docs76%

Data credits: AppWorld leaderboard (Stony Brook NLP). Aggregate scores from the official leaderboard; no results licence asserted. Protected bundles and experiment outputs are not used