Skip to content

Benchmark

AppWorld Test-Challenge: 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

35.6%
Mean
19.7%
Median
30.8%
Std dev
3.4%
10th percentile
85.6%
90th percentile
0.7%
Min
0.7%92.8%
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 docs72.4%
claude-opus-4-7LARA MAS (23715209-d5c9ade3)· AppWorld Leaderboard data docs85.6%
DeepSeekCoderFullCodeRefl (0465049f-bff9fca2)· AppWorld Leaderboard data docs5.8%
DeepSeekCoderPlanExec (871bf9fd-4de66f57)· AppWorld Leaderboard data docs0.7%
DeepSeekCoderReAct (9ae299b8-25c255f5)· AppWorld Leaderboard data docs2.9%
GLM-5.3AtlasHarness (65df675b-ca5e4771)· AppWorld Leaderboard data docs89.5%
GLM-5.3-FlashIBM CUGA (a2db3f1e-2c28b1b2)· AppWorld Leaderboard data docs92.8%
GPT-4 TurboFullCodeRefl (571aa685-a8671c57)· AppWorld Leaderboard data docs12.5%
GPT-4 TurboIPFunCall (08d2ff3e-a0589430)· AppWorld Leaderboard data docs14.6%
GPT-4 TurboPlanExec (c722a232-74b9384b)· AppWorld Leaderboard data docs11%
GPT-4 TurboReAct (1f0055c0-123849bb)· AppWorld Leaderboard data docs17.5%
GPT-4.1IBM CUGA (8a02d49a-0df800d6)· AppWorld Leaderboard data docs57.6%
gpt-4.1-miniLARA MAS (695a4b55-1b68dde4)· AppWorld Leaderboard data docs37.6%
GPT-4oFullCodeRefl (d035381e-2d30c0e3)· AppWorld Leaderboard data docs19.2%
GPT-4oIPFunCall (3a889687-923696fe)· AppWorld Leaderboard data docs18%
GPT-4oPlanExec (ac5b742b-aa0243da)· AppWorld Leaderboard data docs19.7%
GPT-4oReAct (3414a75f-3df6a61f)· AppWorld Leaderboard data docs30.2%
GPT-4oReAct + 2 SetBSR Demos (1c12639a-57a0ac34)· AppWorld Leaderboard data docs38.9%
gpt-5.6-lunakecaipan capybara (1095b768-6513e0c6)· AppWorld Leaderboard data docs73.4%
gptoss-20bRooAGI_agentbuilder_memtune_rl (5acab4bb-badb978f)· AppWorld Leaderboard data docs15.3%
LLaMA3-70BFullCodeRefl (baf95530-cee120fc)· AppWorld Leaderboard data docs7%
LLaMA3-70BPlanExec (77f25849-862250c4)· AppWorld Leaderboard data docs2.4%
LLaMA3-70BReAct (4077db41-07bc5204)· AppWorld Leaderboard data docs3.4%
Qwen2.5-32BLOOP (e5e08b7e-d8e6474f)· AppWorld Leaderboard data docs47.2%
Qwen3-14BAlibaba Cloud ApsaraLab AgentRL (cabab065-0caee3c3)· AppWorld Leaderboard data docs67.6%
Qwen3.5-27BSynthAPI (3bd98253-d4fa6c8e)· AppWorld Leaderboard data docs83.7%

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