Benchmark Almanac

ArenaWebdevPreferences

Qwen-3.5-9B reward model 70.8% tops its leaderboard Gemini 3 Flash 69.1% leads the models we track
Not in model cards

Judges compare pairs of generated applications and predict which application the user preferred. The score measures agreement with user votes.

The researchers collected 6,858 pairs of applications.

Made by
Arena
Released
October 2026
Kind
Web and app front ends
Level
Everyday
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Screens → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Strongest model tested
GPT-5.6 Luna #36 on the capability index
Last model added
July 2026
Scores refreshed
November 20th, 2026
3
Entries in its paper, 2 models we track

Qwen-3.5-9B reward model tops the leaderboard with 70.8% Gemini 3 Flash leads with 69.1%

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Gemini 3 Flash, at 69.1%. This view shows only models we track, each at its best.

  1. 1 Qwen-3.5-9B reward model Full code: Iterative 70.8%
  2. 2 Gemini 3 Flash LLM judge: Iterative 69.1%
  3. 3 GPT-5.6 Luna LLM judge: Code 68.3%

Source: The ArenaWebdevPreferences paper

Source: The ArenaWebdevPreferences paper

#EntryScoreSettingDateSource
1 Qwen-3.5-9B reward model 70.8% Full code: Iterative Official leaderboard
2 Gemini 3 Flash 69.1% LLM judge: Iterative Dec 2025 Official leaderboard
3 GPT-5.6 Luna 68.3% LLM judge: Code Jul 2026 Official leaderboard
#ModelScoreSettingDateSource
1 Gemini 3 Flash 69.1% LLM judge: Iterative Dec 2025 Official leaderboard
2 GPT-5.6 Luna 68.3% LLM judge: Code Jul 2026 Official leaderboard

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Google DeepMind
Benchmark paper Not tested since Oct 2026

Source: The ArenaWebdevPreferences paper

Scores over time

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

Plot with other benchmarks
In the benchmark’s paper
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3 Flash 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3 Flash

Source: Official ArenaWebdevPreferences 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

All 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the benchmark’s paper (3 entries).
  • Note Models scored Only 2 models we track, plus 1 other entry on its leaderboard.
  • Pass Newest score The newest is from July 2026, for GPT-5.6 Luna.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.6 Luna (#36).
  • Pass Headroom The best is 69.1%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 10 rows for models we track and 1 other entry at the last read.
  • Pass Last read November 20th, 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