Benchmarks / DeckBench

Measured by Spring Prompt

DeckBench

Given a finished analysis, which models make a deck a manager could present as it is: true to the numbers, telling the right story, and well designed?

Which AI model makes the best business presentation? Compare the results below, then inspect and download the slide images, read the judges' findings and compare models' decks. DeckBench tests underlying models on the same business briefs; presentation apps add their own templates and editing tools.

Last updated 19 Oct 2026

Results dated
6 Oct 2026 to 19 Oct 2026
Models
24
Unit
% of tasks
Licence
Spring Prompt original
Judge
GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro, by majority · bias check
Runs
1 deck per task, 6 tasks

Recommendation late or wrong: Mercury 3

16 results above zero · % of tasks, lower is better. Choose a model to highlight it.Clear highlight

  1. 9 Claude Sonnet 5.5Anthropic 16.7%
  2. 9 DeepSeek-V4.1-Flash-1017DeepSeek 16.7%
  3. 9 GPT-6 LunaOpenAI 16.7%
  4. 9 Grok 4.7xAI 16.7%
  5. 13 Claude Haiku 5.5Anthropic 33.3%
  6. 13 Gemini 3.8 FlashGoogle 33.3%
  7. 13 MiniMax M3.1 FlashMiniMax 33.3%
  8. 13 Kimi K3Moonshot AI 33.3%
  9. 17 Claude Haiku 4.5Anthropic 50.0%
  10. 17 Gemini 3.1 Pro PreviewGoogle 50.0%
  11. 17 Mistral Medium 3.5Mistral AI 50.0%
  12. 20 DeepSeek-V4-Pro (0423)DeepSeek 66.7%
  13. 20 Muse Spark 1.3Meta 66.7%
  14. 22 Gemini 3.5 Flash-LiteGoogle 83.3%
  15. 22 Mercury 3Inception 83.3%

8 models had none: Claude Fable 5.1, Claude Opus 5.5, GPT-6 Astra, GPT-6 Sol, GPT-6.1 Sol, Gemini 4 Argon, Muse Wonder, Qwen3.8-Max (0902).

Deck rating against cost

What one deck cost at the run date's prices, on a log scale. Up and to the left is better.

Named: the six best and the best for the moneyOther models (hover for names)Best score at each cost

5001,0001,5002,000 $0.001$0.01$0.1$1$10 Cost per deck, US dollars (log scale) Deck rating GPT-6 Astra GPT-6 Sol GPT-6.1 Sol Claude Opus 5.5 Claude Sonnet 5.5 Claude Fable 5.1 GPT-6 Luna

Best per budget

  1. GPT-6 Luna: 1,343 at $0.0087
  2. Claude Sonnet 5.5: 1,408 at $0.15
  3. GPT-6.1 Sol: 1,457 at $0.15
  4. GPT-6 Sol: 1,528 at $0.21
  5. GPT-6 Astra: 1,577 at $1.09

Cheapest first: each model here beats every cheaper one on score.

Full results

DeckBench: recommendation late or wrong, % of tasks, lower is better
#ModelRecommendation late or wrong
% of tasks, lower is better		Deck rating
rating		Presentable decks
% of tasks		Accurate decks
% of tasks		Design quality
% of the maximum		Cost per deck
US dollars
1≈ Qwen3.8-Max (0902)Alibaba
0.0%
0 of 6
1,10033.3%
2 of 6		50.0%
3 of 6		67.0%$0.15
1≈ Claude Fable 5.1Anthropic
0.0%
0 of 6
1,37833.3%
2 of 6		66.7%
4 of 6		66.6%$0.81
1≈ Claude Opus 5.5Anthropic
0.0%
0 of 6
1,43633.3%
2 of 6		83.3%
5 of 6		72.0%$0.26
1≈ Gemini 4 ArgonGoogle
0.0%
0 of 6
1,0260.0%
0 of 6		33.3%
2 of 6		55.2%$0.11
1≈ Muse WonderMeta
0.0%
0 of 6
1,30516.7%
1 of 6		100.0%
6 of 6		67.1%$0.18
1≈ GPT-6.1 SolOpenAI
0.0%
0 of 6
1,45750.0%
3 of 6		100.0%
6 of 6		71.4%$0.15
1≈ GPT-6 AstraOpenAI
0.0%
0 of 6
1,57766.7%
4 of 6		100.0%
6 of 6		88.5%$1.09
1≈ GPT-6 SolOpenAI
0.0%
0 of 6
1,52850.0%
3 of 6		83.3%
5 of 6		87.9%$0.21
9 Claude Sonnet 5.5Anthropic
16.7%
1 of 6
1,40850.0%
3 of 6		66.7%
4 of 6		78.5%$0.15
9 DeepSeek-V4.1-Flash-1017DeepSeek
16.7%
1 of 6
1,1450.0%
0 of 6		33.3%
2 of 6		68.5%$0.0255
9 GPT-6 LunaOpenAI
16.7%
1 of 6
1,3430.0%
0 of 6		50.0%
3 of 6		78.1%$0.0087
9 Grok 4.7xAI
16.7%
1 of 6
1,31833.3%
2 of 6		66.7%
4 of 6		76.4%$0.25
13 Claude Haiku 5.5Anthropic
33.3%
2 of 6
1,19733.3%
2 of 6		33.3%
2 of 6		61.1%$0.0091
13 Gemini 3.8 FlashGoogle
33.3%
2 of 6
8440.0%
0 of 6		16.7%
1 of 6		61.7%$0.0613
13 MiniMax M3.1 FlashMiniMax
33.3%
2 of 6
1,0900.0%
0 of 6		50.0%
3 of 6		46.2%$0
13 Kimi K3Moonshot AI
33.3%
2 of 6
1,2740.0%
0 of 6		50.0%
3 of 6		81.4%$0.45
17 Claude Haiku 4.5Anthropic
50.0%
3 of 6
7020.0%
0 of 6		16.7%
1 of 6		47.6%$0.0461
17 Gemini 3.1 Pro PreviewGoogle
50.0%
3 of 6
78816.7%
1 of 6		50.0%
3 of 6		51.4%$0.14
17 Mistral Medium 3.5Mistral AI
50.0%
3 of 6
6560.0%
0 of 6		0.0%
0 of 6		43.7%$0.0448
20 DeepSeek-V4-Pro (0423)DeepSeek
66.7%
4 of 6
5990.0%
0 of 6		0.0%
0 of 6		40.9%$0.0465
20 Muse Spark 1.3Meta
66.7%
4 of 6
8880.0%
0 of 6		16.7%
1 of 6		56.1%$0.0463
22 Gemini 3.5 Flash-LiteGoogle
83.3%
5 of 6
6740.0%
0 of 6		0.0%
0 of 6		60.2%$0.0126
22 Mercury 3Inception
83.3%
5 of 6
8120.0%
0 of 6		0.0%
0 of 6		58.5%$0.0116
22 Mistral Large 4Mistral AI
83.3%
5 of 6
6370.0%
0 of 6		0.0%
0 of 6		30.5%$0.0148

Swipe the table sideways for more columns.

Ranks follow the score as shown, so equal numbers share a rank. Each model runs at its provider's default reasoning setting, one deck per task. The rating comes from head-to-head comparisons, so it keeps separating models even when several make presentable decks; ranges are 95% intervals from resampling the comparisons.

Progress over time

Recommendation late or wrong, plotted at each model's release date, lower is better. Each model counts once, at its best setting. Scores are today's, not launch day's. Hover or tap a mark for its model.

Best score so far

Every model at its release date. The line moves each time a new release beats every earlier one, and each new best is named where there is room.

0%50%100% 2026AprJulOct Claude Haiku 4.5Kimi K3Qwen3.8-Max (0902)

Best score by lab

Each lab's best so far, stepping whenever one of its releases beats its own earlier models. The 6 labs with the best scores today; 5 others not drawn.

AnthropicGoogleAlibabaMetaOpenAIDeepSeek

-20%0%20%40%60%80% 2026AprJulOct AnthropicGoogleAlibabaMetaOpenAIDeepSeek

See the decks

Every model's deck for every public task, as rendered from the slides it returned. Open a deck to see each slide, the judges' rating and what stopped it being presentable.

Hearthside Coffee: Manchester store investment review · for the board of directors (non-executives and the chief executive)
Compare decks side by side
Title slide of GPT-6 Astra's deck GPT-6 Astra ✓ Presentable Title slide of GPT-6 Sol's deck GPT-6 Sol ✗ 1 issue Title slide of GPT-6.1 Sol's deck GPT-6.1 Sol ✓ Presentable Title slide of Claude Opus 5.5's deck Claude Opus 5.5 ✓ Presentable Title slide of Claude Sonnet 5.5's deck Claude Sonnet 5.5 ✓ Presentable Title slide of Claude Fable 5.1's deck Claude Fable 5.1 ✗ 3 issues Title slide of GPT-6 Luna's deck GPT-6 Luna ✗ 1 issue Title slide of Grok 4.7's deck Grok 4.7 ✓ Presentable Title slide of Muse Wonder's deck Muse Wonder ✗ 2 issues Title slide of Kimi K3's deck Kimi K3 ✗ 5 issues Title slide of Claude Haiku 5.5's deck Claude Haiku 5.5 ✗ 2 issues Title slide of DeepSeek V4.1 Flash 1017's deck DeepSeek V4.1 Flash 1017 ✗ 1 issue Title slide of Qwen3.8 Max (0902)'s deck Qwen3.8 Max (0902) ✗ 2 issues Title slide of minimax-m3.1-flash:free's deck minimax-m3.1-flash:free ✗ 5 issues Title slide of Gemini 4 Argon's deck Gemini 4 Argon ✗ 2 issues Title slide of Muse Spark 1.3's deck Muse Spark 1.3 ✗ 2 issues Title slide of Gemini 3.8 Flash's deck Gemini 3.8 Flash ✗ 2 issues No usable deck Mercury 3 ✗ No deck Title slide of Gemini 3.1 Pro Preview's deck Gemini 3.1 Pro Preview ✗ 2 issues Title slide of Claude Haiku 4.5's deck Claude Haiku 4.5 ✗ 2 issues Title slide of Gemini 3.5 Flash Lite's deck Gemini 3.5 Flash Lite ✗ 4 issues Title slide of Mistral Medium 3.5's deck Mistral Medium 3.5 ✗ 1 issue Title slide of Mistral Large 4's deck Mistral Large 4 ✗ 3 issues Title slide of DeepSeek V4 Pro 0423's deck DeepSeek V4 Pro 0423 ✗ 2 issues No usable deck GLM 5.3 ✗ No deck
Lumio: Starter churn: what to fund in Q1 2026 · for the leadership team (chief executive, CFO, heads of product, sales and customer success)
Compare decks side by side
Title slide of GPT-6 Astra's deck GPT-6 Astra ✓ Presentable Title slide of GPT-6 Sol's deck GPT-6 Sol ✓ Presentable Title slide of GPT-6.1 Sol's deck GPT-6.1 Sol ✗ 2 issues Title slide of Claude Opus 5.5's deck Claude Opus 5.5 ✗ 1 issue Title slide of Claude Sonnet 5.5's deck Claude Sonnet 5.5 ✗ 1 issue Title slide of Claude Fable 5.1's deck Claude Fable 5.1 ✓ Presentable Title slide of GPT-6 Luna's deck GPT-6 Luna ✗ 1 issue Title slide of Grok 4.7's deck Grok 4.7 ✗ 1 issue Title slide of Muse Wonder's deck Muse Wonder ✗ 2 issues Title slide of Kimi K3's deck Kimi K3 ✗ 1 issue No usable deck Claude Haiku 5.5 ✗ No deck Title slide of DeepSeek V4.1 Flash 1017's deck DeepSeek V4.1 Flash 1017 ✗ 2 issues Title slide of Qwen3.8 Max (0902)'s deck Qwen3.8 Max (0902) ✗ 3 issues Title slide of minimax-m3.1-flash:free's deck minimax-m3.1-flash:free ✗ 1 issue Title slide of Gemini 4 Argon's deck Gemini 4 Argon ✗ 3 issues Title slide of Muse Spark 1.3's deck Muse Spark 1.3 ✗ 2 issues Title slide of Gemini 3.8 Flash's deck Gemini 3.8 Flash ✗ 2 issues Title slide of Mercury 3's deck Mercury 3 ✗ 1 issue Title slide of Gemini 3.1 Pro Preview's deck Gemini 3.1 Pro Preview ✗ 1 issue Title slide of Claude Haiku 4.5's deck Claude Haiku 4.5 ✗ 3 issues Title slide of Gemini 3.5 Flash Lite's deck Gemini 3.5 Flash Lite ✗ 3 issues Title slide of Mistral Medium 3.5's deck Mistral Medium 3.5 ✗ 1 issue Title slide of Mistral Large 4's deck Mistral Large 4 ✗ 5 issues Title slide of DeepSeek V4 Pro 0423's deck DeepSeek V4 Pro 0423 ✗ 4 issues No usable deck GLM 5.3 ✗ No deck
Pantry Lane: Paid media: H1 2026 allocation · for the CMO and the finance director
Compare decks side by side
Title slide of GPT-6 Astra's deck GPT-6 Astra ✓ Presentable Title slide of GPT-6 Sol's deck GPT-6 Sol ✗ 1 issue Title slide of GPT-6.1 Sol's deck GPT-6.1 Sol ✗ 1 issue Title slide of Claude Opus 5.5's deck Claude Opus 5.5 ✗ 1 issue Title slide of Claude Sonnet 5.5's deck Claude Sonnet 5.5 ✓ Presentable Title slide of Claude Fable 5.1's deck Claude Fable 5.1 ✗ 3 issues Title slide of GPT-6 Luna's deck GPT-6 Luna ✗ 1 issue Title slide of Grok 4.7's deck Grok 4.7 ✓ Presentable Title slide of Muse Wonder's deck Muse Wonder ✗ 2 issues Title slide of Kimi K3's deck Kimi K3 ✗ 1 issue Title slide of Claude Haiku 5.5's deck Claude Haiku 5.5 ✓ Presentable Title slide of DeepSeek V4.1 Flash 1017's deck DeepSeek V4.1 Flash 1017 ✗ 3 issues Title slide of Qwen3.8 Max (0902)'s deck Qwen3.8 Max (0902) ✓ Presentable Title slide of minimax-m3.1-flash:free's deck minimax-m3.1-flash:free ✗ 2 issues Title slide of Gemini 4 Argon's deck Gemini 4 Argon ✗ 1 issue Title slide of Muse Spark 1.3's deck Muse Spark 1.3 ✗ 1 issue Title slide of Gemini 3.8 Flash's deck Gemini 3.8 Flash ✗ 5 issues Title slide of Mercury 3's deck Mercury 3 ✗ 1 issue Title slide of Gemini 3.1 Pro Preview's deck Gemini 3.1 Pro Preview ✓ Presentable Title slide of Claude Haiku 4.5's deck Claude Haiku 4.5 ✗ 2 issues Title slide of Gemini 3.5 Flash Lite's deck Gemini 3.5 Flash Lite ✗ 4 issues Title slide of Mistral Medium 3.5's deck Mistral Medium 3.5 ✗ 1 issue Title slide of Mistral Large 4's deck Mistral Large 4 ✗ 2 issues Title slide of DeepSeek V4 Pro 0423's deck DeepSeek V4 Pro 0423 ✗ 4 issues No usable deck GLM 5.3 ✗ No deck
Northgate Homes: Responsive repairs: in-house team or renewed contract · for the board of trustees of a housing association, including two tenant board members
Compare decks side by side
Title slide of GPT-6 Astra's deck GPT-6 Astra ✗ 1 issue Title slide of GPT-6 Sol's deck GPT-6 Sol ✓ Presentable Title slide of GPT-6.1 Sol's deck GPT-6.1 Sol ✗ 1 issue Title slide of Claude Opus 5.5's deck Claude Opus 5.5 ✓ Presentable Title slide of Claude Sonnet 5.5's deck Claude Sonnet 5.5 ✓ Presentable Title slide of Claude Fable 5.1's deck Claude Fable 5.1 ✗ 2 issues Title slide of GPT-6 Luna's deck GPT-6 Luna ✗ 1 issue Title slide of Grok 4.7's deck Grok 4.7 ✗ 1 issue Title slide of Muse Wonder's deck Muse Wonder ✓ Presentable Title slide of Kimi K3's deck Kimi K3 ✗ 1 issue Title slide of Claude Haiku 5.5's deck Claude Haiku 5.5 ✗ 1 issue Title slide of DeepSeek V4.1 Flash 1017's deck DeepSeek V4.1 Flash 1017 ✗ 2 issues Title slide of Qwen3.8 Max (0902)'s deck Qwen3.8 Max (0902) ✗ 1 issue Title slide of minimax-m3.1-flash:free's deck minimax-m3.1-flash:free ✗ 2 issues Title slide of Gemini 4 Argon's deck Gemini 4 Argon ✗ 2 issues Title slide of Muse Spark 1.3's deck Muse Spark 1.3 ✗ 4 issues Title slide of Gemini 3.8 Flash's deck Gemini 3.8 Flash ✗ 1 issue Title slide of Mercury 3's deck Mercury 3 ✗ 2 issues Title slide of Gemini 3.1 Pro Preview's deck Gemini 3.1 Pro Preview ✗ 3 issues Title slide of Claude Haiku 4.5's deck Claude Haiku 4.5 ✗ 3 issues Title slide of Gemini 3.5 Flash Lite's deck Gemini 3.5 Flash Lite ✗ 2 issues Title slide of Mistral Medium 3.5's deck Mistral Medium 3.5 ✗ 4 issues Title slide of Mistral Large 4's deck Mistral Large 4 ✗ 5 issues Title slide of DeepSeek V4 Pro 0423's deck DeepSeek V4 Pro 0423 ✗ 4 issues No usable deck GLM 5.3 ✗ No deck
Kestrel Logistics: On-time delivery: what is going wrong and what to fund · for the board and the company's three main investors
Compare decks side by side
Title slide of GPT-6 Astra's deck GPT-6 Astra ✓ Presentable Title slide of GPT-6 Sol's deck GPT-6 Sol ✓ Presentable Title slide of GPT-6.1 Sol's deck GPT-6.1 Sol ✓ Presentable Title slide of Claude Opus 5.5's deck Claude Opus 5.5 ✗ 2 issues Title slide of Claude Sonnet 5.5's deck Claude Sonnet 5.5 ✗ 1 issue Title slide of Claude Fable 5.1's deck Claude Fable 5.1 ✗ 1 issue Title slide of GPT-6 Luna's deck GPT-6 Luna ✗ 1 issue Title slide of Grok 4.7's deck Grok 4.7 ✗ 1 issue Title slide of Muse Wonder's deck Muse Wonder ✗ 2 issues Title slide of Kimi K3's deck Kimi K3 ✗ 1 issue Title slide of Claude Haiku 5.5's deck Claude Haiku 5.5 ✓ Presentable Title slide of DeepSeek V4.1 Flash 1017's deck DeepSeek V4.1 Flash 1017 ✗ 2 issues Title slide of Qwen3.8 Max (0902)'s deck Qwen3.8 Max (0902) ✓ Presentable No usable deck minimax-m3.1-flash:free ✗ No deck Title slide of Gemini 4 Argon's deck Gemini 4 Argon ✗ 3 issues Title slide of Muse Spark 1.3's deck Muse Spark 1.3 ✗ 3 issues Title slide of Gemini 3.8 Flash's deck Gemini 3.8 Flash ✗ 4 issues Title slide of Mercury 3's deck Mercury 3 ✗ 2 issues Title slide of Gemini 3.1 Pro Preview's deck Gemini 3.1 Pro Preview ✗ 3 issues Title slide of Claude Haiku 4.5's deck Claude Haiku 4.5 ✗ 2 issues Title slide of Gemini 3.5 Flash Lite's deck Gemini 3.5 Flash Lite ✗ 1 issue Title slide of Mistral Medium 3.5's deck Mistral Medium 3.5 ✗ 3 issues Title slide of Mistral Large 4's deck Mistral Large 4 ✗ 4 issues Title slide of DeepSeek V4 Pro 0423's deck DeepSeek V4 Pro 0423 ✗ 4 issues No usable deck GLM 5.3 ✗ No deck
Wrenfield Borough Council: Library services: choosing how to save £350,000 · for the council's cabinet (elected members), meeting in public
Compare decks side by side
Title slide of GPT-6 Astra's deck GPT-6 Astra ✗ 1 issue Title slide of GPT-6 Sol's deck GPT-6 Sol ✗ 1 issue Title slide of GPT-6.1 Sol's deck GPT-6.1 Sol ✓ Presentable Title slide of Claude Opus 5.5's deck Claude Opus 5.5 ✗ 2 issues Title slide of Claude Sonnet 5.5's deck Claude Sonnet 5.5 ✗ 2 issues Title slide of Claude Fable 5.1's deck Claude Fable 5.1 ✓ Presentable Title slide of GPT-6 Luna's deck GPT-6 Luna ✗ 1 issue Title slide of Grok 4.7's deck Grok 4.7 ✗ 2 issues Title slide of Muse Wonder's deck Muse Wonder ✗ 1 issue Title slide of Kimi K3's deck Kimi K3 ✗ 1 issue Title slide of Claude Haiku 5.5's deck Claude Haiku 5.5 ✗ 1 issue Title slide of DeepSeek V4.1 Flash 1017's deck DeepSeek V4.1 Flash 1017 ✗ 5 issues Title slide of Qwen3.8 Max (0902)'s deck Qwen3.8 Max (0902) ✗ 3 issues No usable deck minimax-m3.1-flash:free ✗ No deck Title slide of Gemini 4 Argon's deck Gemini 4 Argon ✗ 1 issue Title slide of Muse Spark 1.3's deck Muse Spark 1.3 ✗ 4 issues Title slide of Gemini 3.8 Flash's deck Gemini 3.8 Flash ✗ 2 issues Title slide of Mercury 3's deck Mercury 3 ✗ 1 issue Title slide of Gemini 3.1 Pro Preview's deck Gemini 3.1 Pro Preview ✗ 3 issues Title slide of Claude Haiku 4.5's deck Claude Haiku 4.5 ✗ 4 issues Title slide of Gemini 3.5 Flash Lite's deck Gemini 3.5 Flash Lite ✗ 3 issues Title slide of Mistral Medium 3.5's deck Mistral Medium 3.5 ✗ 3 issues Title slide of Mistral Large 4's deck Mistral Large 4 ✗ 3 issues Title slide of DeepSeek V4 Pro 0423's deck DeepSeek V4 Pro 0423 ✗ 3 issues No usable deck GLM 5.3 ✗ No deck

More from the results

Accurate is not the same as presentable

Share of tasks where the deck said only what the analysis supports, against the share that could also be presented as it is. The gap is layout and design.

AccuratePresentable
Muse Wonder
100.0% · 16.7%
GPT-6.1 Sol
100.0% · 50.0%
GPT-6 Astra
100.0% · 66.7%
Claude Opus 5.5
83.3% · 33.3%
GPT-6 Sol
83.3% · 50.0%
Claude Fable 5.1
66.7% · 33.3%
Claude Sonnet 5.5
66.7% · 50.0%
Grok 4.7
66.7% · 33.3%
Qwen3.8-Max (0902)
50.0% · 33.3%
Gemini 3.1 Pro Preview
50.0% · 16.7%
MiniMax M3.1 Flash
50.0% · 0.0%
Kimi K3
50.0% · 0.0%
GPT-6 Luna
50.0% · 0.0%
Claude Haiku 5.5
33.3% · 33.3%
DeepSeek-V4.1-Flash-1017
33.3% · 0.0%
Gemini 4 Argon
33.3% · 0.0%
Claude Haiku 4.5
16.7% · 0.0%
Gemini 3.8 Flash
16.7% · 0.0%
Muse Spark 1.3
16.7% · 0.0%
DeepSeek-V4-Pro (0423)
0.0% · 0.0%
Gemini 3.5 Flash-Lite
0.0% · 0.0%
Mercury 3
0.0% · 0.0%
Mistral Large 4
0.0% · 0.0%
Mistral Medium 3.5
0.0% · 0.0%

Why decks are not presentable

Share of tasks failing each check. A deck can fail several at once; any one means it needs fixing before it is shown.

ModelLayout defectsSlides needing workDraft figure quotedUnsupported numbersUnsupported claimsCaveat droppedMisleading metricFindings missingRecommendation
GPT-6 AstraOpenAI0.0%33.3%0.0%0.0%0.0%0.0%0.0%0.0%0.0%
GPT-6 SolOpenAI33.3%0.0%0.0%0.0%0.0%0.0%0.0%16.7%0.0%
GPT-6.1 SolOpenAI16.7%50.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%
Claude Opus 5.5Anthropic16.7%66.7%0.0%0.0%0.0%16.7%0.0%0.0%0.0%
Claude Sonnet 5.5Anthropic0.0%33.3%0.0%0.0%0.0%16.7%0.0%0.0%16.7%
Claude Fable 5.1Anthropic66.7%50.0%33.3%0.0%0.0%0.0%0.0%0.0%0.0%
GPT-6 LunaOpenAI33.3%16.7%33.3%0.0%0.0%0.0%0.0%0.0%16.7%
Grok 4.7xAI16.7%16.7%16.7%0.0%0.0%16.7%0.0%0.0%16.7%
Muse WonderMeta83.3%66.7%0.0%0.0%0.0%0.0%0.0%0.0%0.0%
Kimi K3Moonshot AI66.7%16.7%0.0%0.0%33.3%0.0%0.0%16.7%33.3%
Claude Haiku 5.5Anthropic33.3%16.7%33.3%16.7%33.3%16.7%16.7%16.7%33.3%
DeepSeek-V4.1-Flash-1017DeepSeek83.3%66.7%50.0%0.0%16.7%16.7%0.0%0.0%16.7%
Qwen3.8-Max (0902)Alibaba33.3%50.0%33.3%0.0%16.7%16.7%0.0%0.0%0.0%
MiniMax M3.1 FlashMiniMax100.0%83.3%33.3%50.0%50.0%33.3%33.3%50.0%33.3%
Gemini 4 ArgonGoogle66.7%66.7%50.0%0.0%16.7%0.0%0.0%0.0%0.0%
Muse Spark 1.3Meta100.0%50.0%16.7%0.0%16.7%16.7%0.0%0.0%66.7%
Gemini 3.8 FlashGoogle83.3%33.3%50.0%16.7%50.0%0.0%0.0%0.0%33.3%
Mercury 3Inception33.3%33.3%33.3%16.7%16.7%16.7%16.7%16.7%83.3%
Gemini 3.1 Pro PreviewGoogle50.0%83.3%0.0%0.0%0.0%16.7%0.0%0.0%50.0%
Claude Haiku 4.5Anthropic50.0%66.7%50.0%0.0%33.3%0.0%0.0%16.7%50.0%
Gemini 3.5 Flash-LiteGoogle66.7%16.7%50.0%0.0%16.7%16.7%0.0%33.3%83.3%
Mistral Medium 3.5Mistral AI33.3%50.0%50.0%0.0%33.3%0.0%0.0%0.0%50.0%
Mistral Large 4Mistral AI66.7%100.0%16.7%33.3%50.0%16.7%0.0%0.0%83.3%
DeepSeek-V4-Pro (0423)DeepSeek66.7%83.3%33.3%0.0%66.7%16.7%0.0%16.7%66.7%

Every task, every model

Whether each deck could be presented as it is, and if not, how many kinds of issue it had. Each square opens the deck.

ModelHearthside CoffeeLumioPantry LaneNorthgate HomesKestrel LogisticsWrenfield Borough CouncilPresentable
GPT-6 Astra✓✓✓1✓14 of 6
GPT-6 Sol1✓1✓✓13 of 6
GPT-6.1 Sol✓211✓✓3 of 6
Claude Sonnet 5.5✓1✓✓123 of 6
Claude Opus 5.5✓11✓222 of 6
Claude Fable 5.13✓321✓2 of 6
Grok 4.7✓1✓1122 of 6
Claude Haiku 5.52–✓1✓12 of 6
Qwen3.8 Max (0902)23✓1✓32 of 6
Muse Wonder222✓211 of 6
Gemini 3.1 Pro Preview21✓3331 of 6
GPT-6 Luna1111110 of 6
Kimi K35111110 of 6
DeepSeek V4.1 Flash 10171232250 of 6
minimax-m3.1-flash:free5122––0 of 6
Gemini 4 Argon2312310 of 6
Muse Spark 1.32214340 of 6
Gemini 3.8 Flash2251420 of 6
Mercury 3–112210 of 6
Claude Haiku 4.52323240 of 6
Gemini 3.5 Flash Lite4342130 of 6
Mistral Medium 3.51114330 of 6
Mistral Large 43525430 of 6
DeepSeek V4 Pro 04232444430 of 6
GLM 5.3––––––0 of 6

✓ presentable as it is · a number: the kinds of issue that stopped it · – no usable deck.

How DeckBench works

  1. 1

    The analysis

    An invented company's memo with findings, a recommendation and caveats, two to four data tables, the audience, the decision and a brand kit.

  2. 2

    The deck

    The model returns every slide as positioned text, shapes, native charts and tables. We render it to PowerPoint and to images.

  3. 3

    The checks

    Code checks the layout and traces every number to the analysis; three judges check the story and rate each slide's design.

  4. 4

    Head to head

    On each task the judges compare decks in pairs and pick the one they would rather present. The votes become the rating.

The task

Turning an analysis into slides is one of the most common things people ask AI to do at work, and the risk is not ugly slides: it is a number that is not in the analysis, a caveat that falls off, or a draft figure that ends up in front of the board.

Each task is an invented company's analysis written for the benchmark, with planted traps: a caveat that must travel with its number, a figure the memo says not to quote, a metric it warns is misleading, and a finding that contradicts a senior stakeholder's view.

What makes a deck presentable

Layout
Nothing off the slide or colliding, body text at least 12 points, and text readable against its background. Text wraps from real font metrics, so overflow is measured.
Numbers
Every number on a slide, in a chart or in a table must come from the analysis; numbers code cannot match go to the judges.
Story
The key findings, the recommendation within the first two slides, caveats with their numbers, and no misleading metric used as evidence.
Design
Each slide rated polished, acceptable or needs work. A presentable deck has no slide that needs work.

Why a rating as well

Presentable is a bar: once the best models clear it on every task, it stops telling them apart. The head-to-head rating does not run out of room, because a deck only has to be preferred to another one. Each deck meets about six others per task; a 400-point gap means the higher-rated model's deck is preferred about ten times out of eleven.

What it measures

  • Every number traced to the analysis
  • The story: findings, recommendation and caveats
  • Layout and legibility, checked by code
  • Design quality, judged slide by slide

What it does not measure

  • Your templates, data or brand guidelines
  • Decks built in PowerPoint or Google Slides by an agent clicking through the app
  • Speaking notes or delivery

Method

  • Layout, legibility and numbers are checked by code
  • Three judges from three vendors decide each judged check by majority
  • No judge compares decks from its own vendor
  • Two tasks are held back so the set can be refreshed

Checking the judge

A single judge was the strictest of three on unsupported claims (it flagged 24 material claims in a pilot where the other two flagged 12 and 1), and its own model led the pilot. So every judged check is decided by a majority of three judges: GPT-6.1 Sol, Claude Opus 5.5 and Gemini 3.1 Pro. In the pilot they agreed with each other on 97 to 100% of the checks on findings, numbers, recommendations and misleading metrics, and on 89 to 90% of decisions about whether a slide needs work. In head-to-head comparisons a judge never votes on a pair that includes its own vendor's deck. A sample of human ratings is still to come.

Failures

A model that made no usable deck on any task has no rating, so it is listed here instead of ranked. They are listed here so you can see why.

Not ranked

  • GLM 5.3: no usable deck: reply cut off at the token limit on 6 of 6 tasks