Skip to content

Benchmark

AppWorld Test-Challenge Level 1: 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

55.2%
Mean
55.6%
Median
28.9%
Std dev
16.7%
10th percentile
93.1%
90th percentile
4.2%
Min
4.2%95.8%
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 docs76.4%
claude-opus-4-7LARA MAS (23715209-d5c9ade3)· AppWorld Leaderboard data docs91.7%
DeepSeekCoderFullCodeRefl (0465049f-bff9fca2)· AppWorld Leaderboard data docs29.2%
DeepSeekCoderPlanExec (871bf9fd-4de66f57)· AppWorld Leaderboard data docs4.2%
DeepSeekCoderReAct (9ae299b8-25c255f5)· AppWorld Leaderboard data docs16.7%
GLM-5.3AtlasHarness (65df675b-ca5e4771)· AppWorld Leaderboard data docs93.1%
GLM-5.3-FlashIBM CUGA (a2db3f1e-2c28b1b2)· AppWorld Leaderboard data docs93.1%
GPT-4 TurboFullCodeRefl (571aa685-a8671c57)· AppWorld Leaderboard data docs47.2%
GPT-4 TurboIPFunCall (08d2ff3e-a0589430)· AppWorld Leaderboard data docs22.2%
GPT-4 TurboPlanExec (c722a232-74b9384b)· AppWorld Leaderboard data docs36.1%
GPT-4 TurboReAct (1f0055c0-123849bb)· AppWorld Leaderboard data docs47.2%
GPT-4.1IBM CUGA (8a02d49a-0df800d6)· AppWorld Leaderboard data docs91.7%
gpt-4.1-miniLARA MAS (695a4b55-1b68dde4)· AppWorld Leaderboard data docs72.2%
GPT-4oFullCodeRefl (d035381e-2d30c0e3)· AppWorld Leaderboard data docs54.2%
GPT-4oIPFunCall (3a889687-923696fe)· AppWorld Leaderboard data docs27.8%
GPT-4oPlanExec (ac5b742b-aa0243da)· AppWorld Leaderboard data docs50%
GPT-4oReAct (3414a75f-3df6a61f)· AppWorld Leaderboard data docs58.3%
GPT-4oReAct + 2 SetBSR Demos (1c12639a-57a0ac34)· AppWorld Leaderboard data docs68.1%
gpt-5.6-lunakecaipan capybara (1095b768-6513e0c6)· AppWorld Leaderboard data docs83.3%
gptoss-20bRooAGI_agentbuilder_memtune_rl (5acab4bb-badb978f)· AppWorld Leaderboard data docs55.6%
LLaMA3-70BFullCodeRefl (baf95530-cee120fc)· AppWorld Leaderboard data docs34.7%
LLaMA3-70BPlanExec (77f25849-862250c4)· AppWorld Leaderboard data docs11.1%
LLaMA3-70BReAct (4077db41-07bc5204)· AppWorld Leaderboard data docs12.5%
Qwen2.5-32BLOOP (e5e08b7e-d8e6474f)· AppWorld Leaderboard data docs73.6%
Qwen3-14BAlibaba Cloud ApsaraLab AgentRL (cabab065-0caee3c3)· AppWorld Leaderboard data docs88.9%
Qwen3.5-27BSynthAPI (3bd98253-d4fa6c8e)· AppWorld Leaderboard data docs95.8%

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