Skip to content

Benchmark

AppWorld Test-Challenge: Scenario 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

26.3%
Mean
12.2%
Median
28.1%
Std dev
0.7%
10th percentile
77%
90th percentile
0%
Min
0%84.2%
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 docs64.8%
claude-opus-4-7LARA MAS (23715209-d5c9ade3)· AppWorld Leaderboard data docs77%
DeepSeekCoderFullCodeRefl (0465049f-bff9fca2)· AppWorld Leaderboard data docs2.9%
DeepSeekCoderPlanExec (871bf9fd-4de66f57)· AppWorld Leaderboard data docs0%
DeepSeekCoderReAct (9ae299b8-25c255f5)· AppWorld Leaderboard data docs0.7%
GLM-5.3AtlasHarness (65df675b-ca5e4771)· AppWorld Leaderboard data docs78.4%
GLM-5.3-FlashIBM CUGA (a2db3f1e-2c28b1b2)· AppWorld Leaderboard data docs84.2%
GPT-4 TurboFullCodeRefl (571aa685-a8671c57)· AppWorld Leaderboard data docs7.2%
GPT-4 TurboIPFunCall (08d2ff3e-a0589430)· AppWorld Leaderboard data docs9.3%
GPT-4 TurboPlanExec (c722a232-74b9384b)· AppWorld Leaderboard data docs3.6%
GPT-4 TurboReAct (1f0055c0-123849bb)· AppWorld Leaderboard data docs5.8%
GPT-4.1IBM CUGA (8a02d49a-0df800d6)· AppWorld Leaderboard data docs48.2%
gpt-4.1-miniLARA MAS (695a4b55-1b68dde4)· AppWorld Leaderboard data docs20.1%
GPT-4oFullCodeRefl (d035381e-2d30c0e3)· AppWorld Leaderboard data docs12.2%
GPT-4oIPFunCall (3a889687-923696fe)· AppWorld Leaderboard data docs10.1%
GPT-4oPlanExec (ac5b742b-aa0243da)· AppWorld Leaderboard data docs7.9%
GPT-4oReAct (3414a75f-3df6a61f)· AppWorld Leaderboard data docs13%
GPT-4oReAct + 2 SetBSR Demos (1c12639a-57a0ac34)· AppWorld Leaderboard data docs23%
gpt-5.6-lunakecaipan capybara (1095b768-6513e0c6)· AppWorld Leaderboard data docs52.5%
gptoss-20bRooAGI_agentbuilder_memtune_rl (5acab4bb-badb978f)· AppWorld Leaderboard data docs7.9%
LLaMA3-70BFullCodeRefl (baf95530-cee120fc)· AppWorld Leaderboard data docs4.3%
LLaMA3-70BPlanExec (77f25849-862250c4)· AppWorld Leaderboard data docs0.7%
LLaMA3-70BReAct (4077db41-07bc5204)· AppWorld Leaderboard data docs0%
Qwen2.5-32BLOOP (e5e08b7e-d8e6474f)· AppWorld Leaderboard data docs28.8%
Qwen3-14BAlibaba Cloud ApsaraLab AgentRL (cabab065-0caee3c3)· AppWorld Leaderboard data docs50.4%
Qwen3.5-27BSynthAPI (3bd98253-d4fa6c8e)· AppWorld Leaderboard data docs71.2%

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