Skip to content

Benchmark

AppWorld Test-Challenge Level 3: 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

30%
Mean
10.3%
Median
33.3%
Std dev
0%
10th percentile
85.1%
90th percentile
0%
Min
0%94.4%
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 docs72.8%
claude-opus-4-7LARA MAS (23715209-d5c9ade3)· AppWorld Leaderboard data docs84.1%
DeepSeekCoderFullCodeRefl (0465049f-bff9fca2)· AppWorld Leaderboard data docs0%
DeepSeekCoderPlanExec (871bf9fd-4de66f57)· AppWorld Leaderboard data docs0%
DeepSeekCoderReAct (9ae299b8-25c255f5)· AppWorld Leaderboard data docs0%
GLM-5.3AtlasHarness (65df675b-ca5e4771)· AppWorld Leaderboard data docs90.3%
GLM-5.3-FlashIBM CUGA (a2db3f1e-2c28b1b2)· AppWorld Leaderboard data docs94.4%
GPT-4 TurboFullCodeRefl (571aa685-a8671c57)· AppWorld Leaderboard data docs2.1%
GPT-4 TurboIPFunCall (08d2ff3e-a0589430)· AppWorld Leaderboard data docs8.2%
GPT-4 TurboPlanExec (c722a232-74b9384b)· AppWorld Leaderboard data docs2.1%
GPT-4 TurboReAct (1f0055c0-123849bb)· AppWorld Leaderboard data docs7.2%
GPT-4.1IBM CUGA (8a02d49a-0df800d6)· AppWorld Leaderboard data docs44.1%
gpt-4.1-miniLARA MAS (695a4b55-1b68dde4)· AppWorld Leaderboard data docs26.2%
GPT-4oFullCodeRefl (d035381e-2d30c0e3)· AppWorld Leaderboard data docs6.7%
GPT-4oIPFunCall (3a889687-923696fe)· AppWorld Leaderboard data docs9.7%
GPT-4oPlanExec (ac5b742b-aa0243da)· AppWorld Leaderboard data docs10.3%
GPT-4oReAct (3414a75f-3df6a61f)· AppWorld Leaderboard data docs21%
GPT-4oReAct + 2 SetBSR Demos (1c12639a-57a0ac34)· AppWorld Leaderboard data docs31.8%
gpt-5.6-lunakecaipan capybara (1095b768-6513e0c6)· AppWorld Leaderboard data docs73.3%
gptoss-20bRooAGI_agentbuilder_memtune_rl (5acab4bb-badb978f)· AppWorld Leaderboard data docs3.6%
LLaMA3-70BFullCodeRefl (baf95530-cee120fc)· AppWorld Leaderboard data docs1.5%
LLaMA3-70BPlanExec (77f25849-862250c4)· AppWorld Leaderboard data docs0.5%
LLaMA3-70BReAct (4077db41-07bc5204)· AppWorld Leaderboard data docs0%
Qwen2.5-32BLOOP (e5e08b7e-d8e6474f)· AppWorld Leaderboard data docs42.6%
Qwen3-14BAlibaba Cloud ApsaraLab AgentRL (cabab065-0caee3c3)· AppWorld Leaderboard data docs63.6%
Qwen3.5-27BSynthAPI (3bd98253-d4fa6c8e)· AppWorld Leaderboard data docs85.1%

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