Benchmark Almanac

App-Bench

Orchids 76.8% tops its leaderboard Claude Opus 4.5 67.5% leads the models we track
Not in model cards

Coding agents build six full-stack web applications from a single prompt. Two developers grade implemented features against 151 rubric items. The score uses each tool’s best result from three attempts per application.

The benchmark compares complete app-building tools, including their model and platform constraints. The published comparison uses December 2025 configurations.

Made by
AfterQuery
Released
December 2025
Kind
Building whole projects
Level
University
Scored
Graded by people
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Code
Interaction
Agentic
Environment
Terminal
Refresh
One fixed set
Run by
Anyone
In views
Agents
Strongest model tested
Claude Opus 4.5 #62 on the capability index
Last model added
November 2025
Scores refreshed
November 18th, 2026
10
Entries on its own leaderboard, 2 models we track

Orchids tops the leaderboard with 76.8% Claude Opus 4.5 leads with 67.5%

Best score per entry from the benchmark's own leaderboard, read on November 18th, 2026. Where a model ran on its own with several settings, we keep its best. Agents other companies built around a model get a pale shade of that model's lab color. Gray marks blends and entries not in our model list, such as fine-tunes and older models. None of these count toward leaders. The best model we track is Claude Opus 4.5, at 67.5%. Agents built around a model, such as Google AI Studio + Gemini 3 Pro, are their own entries. We don't count them as the model's score. This view shows only models we track, each at its best.

  1. 1 Orchids best of three builds 76.8%
  2. 2 Claude Opus 4.5 best of three builds 67.5%
  3. 3 v0 best of three builds 64.9%
  4. 4 Bolt best of three builds 53.6%
  5. 5 Google AI Studio + Gemini 3 Pro best of three builds 50.3%
  6. 6 GPT-5.1-Codex-Max best of three builds 38.4%
  7. 7 Replit best of three builds 35.1%
  8. 8 Cursor + Composer 1 best of three builds 27.8%
  9. 9 Lovable best of three builds 25.8%
  10. 10 Gemini 2.5 Pro (snapshot unspecified) best of three builds 0%

Source: Official App-Bench leaderboard

#EntryScoreSettingDateSource
1 Orchids 76.8% best of three builds Official leaderboard
2 Claude Opus 4.5 67.5% best of three builds, Claude Code Nov 2025 Official leaderboard
3 v0 64.9% best of three builds Official leaderboard
4 Bolt 53.6% best of three builds Official leaderboard
5 Google AI Studio + Gemini 3 Pro 50.3% best of three builds Nov 2025 Official leaderboard
6 GPT-5.1-Codex-Max 38.4% best of three builds, Codex Nov 2025 Official leaderboard
7 Replit 35.1% best of three builds Official leaderboard
8 Cursor + Composer 1 27.8% best of three builds Oct 2025 Official leaderboard
9 Lovable 25.8% best of three builds Official leaderboard
10 Gemini 2.5 Pro (snapshot unspecified) 0% best of three builds, Gemini CLI Official leaderboard
#ModelScoreSettingDateSource
1 Claude Opus 4.5 67.5% best of three builds, Claude Code Nov 2025 Official leaderboard
2 GPT-5.1-Codex-Max 38.4% best of three builds, Codex Nov 2025 Official leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 4.5 GPT-5.1-Codex-Max 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 4.5 GPT-5.1-Codex-Max

Source: Official App-Bench leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Coding

All 523 in Coding
  • SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
  • LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests. DeepSeek-V4-Pro 93.5%
  • SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
  • DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
  • SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases. Seed2.1 Pro 59.8%
  • Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance. DeepSeek V4.1 Flash 3,471

Data audit

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 12 months ago, in November 2025, for Claude Opus 4.5. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (10 entries).
  • Pass Models scored 2 models we track, plus 8 other entries on its leaderboard.
  • Warning Newest score The newest came 12 months ago, in November 2025, for Claude Opus 4.5.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Claude Opus 4.5 (#62).
  • Pass Headroom The best is 67.5%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on appbench.ai: 4 rows for models we track and 6 other entries at the last read.
  • Pass Last read November 18th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check