Benchmark Almanac

PlotPoints Round 2: Multi-Turn Arena

Claude Opus 4.7 1,575 leads since April 2026
Not in model cards
Version 2

Models roleplay for twelve turns on the same starting scenes. Readers compare two complete conversations with the model names hidden.

The maker’s final leaderboard ranks 20 models from 1,943 votes. A twenty-first model has no human rating.

Released
April 2026
Kind
Creative writing
Level
Everyday
Scored
Ranked by votes
Answers
Open-ended
Tasks
All public
Measures
Preference
Language
English
In and out
Text → Text
Interaction
Multi-turn
Items from
Live play or votes
Run by
A host
Strongest model tested
Claude Opus 4.7 #31 on the capability index
Last model added
April 2026
Scores refreshed
November 20th, 2026
20
Entries on its own leaderboard, 19 models we track

Claude Opus 4.7 leads with 1,575 among models released by April 2026 Claude Opus 4.7 leads with 1,575 among models released by April 2026

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. 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. This view shows only models we track, each at its best.

The newest model with a score here came out in April 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

Bars start at 1,400.

  1. 1 Claude Opus 4.7 1,575
  2. 2 DeepSeek-V4-Pro 1,546
  3. 3 Gemini 3.1 Flash-Lite 1,533
  4. 4 Claude Opus 4.6 1,532
  5. 5 GPT-4.1 1,525
  6. 6 Mistral Small Creative 1,519
  7. 7 Gemini 3.1 Pro 1,513
  8. 8 GLM-4.7 1,510
  9. 9 Kimi K2.6 1,505
  10. 10 DeepSeek-V4-Flash 1,492
  11. 11 Kimi K2.5 1,489
  12. 12 MiniMax-M2.7 1,487
  13. 13 Claude Sonnet 4.5 1,479
  14. 14 DeepSeek-V3.2 1,472
  15. 15 Gemma 4 26B A4B 1,466
  16. 16 Llama 4 Maverick 1,458
  17. 17 GLM-5.1 1,446
  18. 18 Grok 4.1 1,432
  19. 19 Gemini 2.5 Flash Sep 2025 1,418
  20. 20 Qwen3.5-Flash 1,412

Source: Official PlotPoints Round 2: Multi-Turn Arena leaderboard

#EntryScore ±SettingDateSource
1 Claude Opus 4.7 1,575 ±48 Apr 2026 Official leaderboard
2 DeepSeek-V4-Pro 1,546 ±49 Apr 2026 Official leaderboard
3 Gemini 3.1 Flash-Lite 1,533 ±47 Mar 2026 Official leaderboard
4 Claude Opus 4.6 1,532 ±44 Feb 2026 Official leaderboard
5 GPT-4.1 1,525 ±43 Apr 2025 Official leaderboard
6 Mistral Small Creative open weights 1,519 ±46 Official leaderboard
7 Gemini 3.1 Pro 1,513 ±46 Feb 2026 Official leaderboard
8 GLM-4.7 1,510 ±44 Dec 2025 Official leaderboard
9 Kimi K2.6 1,505 ±45 Apr 2026 Official leaderboard
10 DeepSeek-V4-Flash 1,492 ±48 Apr 2026 Official leaderboard
11 Kimi K2.5 1,489 ±48 Jan 2026 Official leaderboard
12 MiniMax-M2.7 1,487 ±45 Mar 2026 Official leaderboard
13 Claude Sonnet 4.5 1,479 ±45 Sep 2025 Official leaderboard
14 DeepSeek-V3.2 1,472 ±46 Dec 2025 Official leaderboard
15 Gemma 4 26B A4B 1,466 ±44 Apr 2026 Official leaderboard
16 Llama 4 Maverick 1,458 ±45 Apr 2025 Official leaderboard
17 GLM-5.1 1,446 ±46 Apr 2026 Official leaderboard
18 Grok 4.1 1,432 ±44 Nov 2025 Official leaderboard
19 Gemini 2.5 Flash Sep 2025 1,418 ±45 Sep 2025 Official leaderboard
20 Qwen3.5-Flash 1,412 ±47 Feb 2026 Official leaderboard
#ModelScore ±SettingDateSource
1 Claude Opus 4.7 1,575 ±48 Apr 2026 Official leaderboard
2 DeepSeek-V4-Pro 1,546 ±49 Apr 2026 Official leaderboard
3 Gemini 3.1 Flash-Lite 1,533 ±47 Mar 2026 Official leaderboard
4 Claude Opus 4.6 1,532 ±44 Feb 2026 Official leaderboard
5 GPT-4.1 1,525 ±43 Apr 2025 Official leaderboard
6 Gemini 3.1 Pro 1,513 ±46 Feb 2026 Official leaderboard
7 GLM-4.7 1,510 ±44 Dec 2025 Official leaderboard
8 Kimi K2.6 1,505 ±45 Apr 2026 Official leaderboard
9 DeepSeek-V4-Flash 1,492 ±48 Apr 2026 Official leaderboard
10 Kimi K2.5 1,489 ±48 Jan 2026 Official leaderboard
11 MiniMax-M2.7 1,487 ±45 Mar 2026 Official leaderboard
12 Claude Sonnet 4.5 1,479 ±45 Sep 2025 Official leaderboard
13 DeepSeek-V3.2 1,472 ±46 Dec 2025 Official leaderboard
14 Gemma 4 26B A4B 1,466 ±44 Apr 2026 Official leaderboard
15 Llama 4 Maverick 1,458 ±45 Apr 2025 Official leaderboard
16 GLM-5.1 1,446 ±46 Apr 2026 Official leaderboard
17 Grok 4.1 1,432 ±44 Nov 2025 Official leaderboard
18 Gemini 2.5 Flash Sep 2025 1,418 ±45 Sep 2025 Official leaderboard
19 Qwen3.5-Flash 1,412 ±47 Feb 2026 Official leaderboard

2 models have held number one on the official leaderboard since its release in April 2026

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

Google DeepMind Anthropic
Own leaderboard Replaced by 3: Multi-Turn Arena, Jun 2026

Source: Official PlotPoints Round 2: Multi-Turn Arena leaderboard

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
On the benchmark’s own leaderboard
1,400 1,450 1,500 1,550 1,600 2025 2026 1,400 1,450 1,500 1,550 1,600 2025 2026

Source: Official PlotPoints Round 2: Multi-Turn Arena 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.

Versions

More in Writing and taste

All 65 in Writing and taste
  • Design Bench (Moonshot) In use Models make graphic designs from direct prompts or few-shot reference images. An automated grader rates the results as Elo scores. Kimi K3.1 Pro 957
  • WritingBench Fading Models complete 1,000 real-world writing requests across 6 domains and 100 subdomains. A judge model scores each reply on five query-specific criteria. GPT-5 83.9
  • Creative Writing (MiniMax in-house) Only in system cards Models write creative pieces for real Hailuo AI user requests. Automatic assessments score the writing. MiniMax-Text-01 81.3%
  • DeepSeek Chinese functional writing evaluation Only in system cards Models answer 3,170 everyday Chinese writing prompts such as reports, emails and speeches. Pairwise judging against a baseline model gives a win rate. DeepSeek-V4-Pro 62.6%
  • Slide generation human evaluation (Z.ai) Only in system cards People compare HTML slide decks from two models on content, layout and visual quality. Metric is overall win rate. GLM-5 67.5%
  • Task-dependent output homogenization evaluation Only in system cards Models answer prompts several times. A judge scores functional diversity where it is wanted and consistency where one answer is right. Muse Spark 0.21

Data audit

All 8 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
  • Pass Independent runs Scores from the official leaderboard (20 entries).
  • Pass Models scored 19 models we track, plus 1 other entry on its leaderboard.
  • Note Newest score The newest came 7 months ago, in April 2026, for DeepSeek-V4-Flash. That’s expected for a test replaced by PlotPoints Round 3: Multi-Turn Arena.
  • Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is Claude Opus 4.7 (#31).
  • Note Headroom Scored as an Elo rating, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on plotlightstudios.com: 19 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