← All questions

Pick a random number between 1 and 6

Overview

4 83.9% 95 of 106 models agree
4 84%
3 11%
4 · 83.9%
3 · 10.8%
5 · 2.7%
6 · 1.7%
Other answers · 0.9% + 3 answers
  • 2 0.4%
  • 1 0.4%
  • other 0.1%

By country of origin

Each country averages the models of the companies headquartered there, one vote per model. Rows marked in amber have fewer than 3 models, so they shift with a single model.
United States
62 models
4 89%
China
35 models
4 85%
France
4 models
4 44%
3 56%
South Korea
4 models
4 33%
3 25%
5 13%
6 30%
Japan
1 model
4 100%

By company

Each company averages its own models, one vote per model.
OpenAI
16 models
4 98%
Anthropic
12 models
4 99%
Qwen
9 models
4 87%
Google
8 models
4 93%
xAI
6 models
4 77%
3 19%
Z.ai
6 models
4 88%
3 11%
MiniMax
5 models
4 94%
MoonshotAI
5 models
4 89%
DeepSeek
4 models
4 60%
3 21%
Mistral
4 models
4 44%
3 56%
Upstage
4 models
4 33%
3 25%
5 13%
6 30%
Xiaomi
4 models
4 95%
Cloudflare
3 models
4 100%
Inception
3 models
4 52%
3 40%
Liquid AI
3 models
4 51%
3 49%
Meta
3 models
4 95%
+ 9 with fewer than 3 models
NVIDIA
2 models
4 100%
TypeSafe
2 models
4 45%
3 35%
Other answers 20%
IBM
1 model
4 100%
Jared Palmer
1 model
4 70%
3 30%
Microsoft
1 model
4 100%
Sakana
1 model
4 100%
StepFun
1 model
4 45%
3 10%
5 45%
Tencent
1 model
4 85%
3 10%
ThinkingMachines
1 model
4 100%

By model release quarter

Each quarter averages only the models released in that quarter. 7 models excluded (no release date on record).
0%50%100% 4 Q3 2024n=2 Q2 2025n=3 Q3 n=4 Q4 n=5 Q1 2026n=18 Q2 n=29 Q3 n=21 Q4 n=17 2024 Q3 · 2 models 4: 60% 3: 40% 2025 Q2 · 3 models 4: 69% 3: 31% 2025 Q3 · 4 models 4: 94% 3: 4% 5: 2% 2025 Q4 · 5 models 4: 81% 3: 14% 5: 1% Other answers: 4% 2026 Q1 · 18 models 4: 85% 3: 8% 5: 4% 6: 2% Other answers: 1% 2026 Q2 · 29 models 4: 89% 3: 6% 5: 4% 2026 Q3 · 21 models 4: 74% 3: 17% 5: 3% 6: 5% Other answers: 2% 2026 Q4 · 17 models 4: 90% 3: 6% 5: 1% 6: 2%
4
3
5
6
Other answers
+ Show data
Quarter Models 4356Other answers
2024 Q3 2 60%40%0%0%0%
2025 Q2 3 69%31%0%0%0%
2025 Q3 4 94%4%2%0%0%
2025 Q4 5 81%14%1%0%4%
2026 Q1 18 85%8%4%2%1%
2026 Q2 29 89%6%4%0%0%
2026 Q3 21 74%17%3%5%2%
2026 Q4 17 90%6%1%2%0%

Outliers

Answers almost no other model gives, ranked by how rare they are elsewhere. Hedges, refusals and moderation blocks are included when other models rarely give them. Most models said “4”. Prompt numbers refer to the user messages listed below.
6 0.8% among other models upstage/solar-decide 10/10 runs
+ 10 responses

upstage/solar-decide

Prompt 1

{"1":0.257467,"2":0.039484,"3":0.044741,"4":0.039484,"5":0.073766,"6":0.545058}

Prompt 1

{"1":0.240654,"2":0.036906,"3":0.047388,"4":0.036906,"5":0.060847,"6":0.5773}

Prompt 1

{"1":0.250885,"2":0.033954,"3":0.029964,"4":0.033954,"5":0.049402,"6":0.601842}

Prompt 1

{"1":0.182424,"2":0.035921,"3":0.046124,"4":0.0317,"5":0.06711,"6":0.636721}

Prompt 2

{"1":0.159831,"2":0.027774,"3":0.058798,"4":0.024511,"5":0.096942,"6":0.632143}

Prompt 2

{"1":0.148356,"2":0.022751,"3":0.061844,"4":0.022751,"5":0.079409,"6":0.664887}

Prompt 2

{"1":0.122372,"2":0.024096,"3":0.051012,"4":0.024096,"5":0.074222,"6":0.704201}

Prompt 3

{"1":0.265834,"2":0.040767,"3":0.067213,"4":0.031749,"5":0.097795,"6":0.496642}

Prompt 3

{"1":0.285506,"2":0.038639,"3":0.081799,"4":0.043784,"5":0.134864,"6":0.415409}

Prompt 3

{"1":0.321092,"2":0.043455,"3":0.104243,"4":0.049241,"5":0.118123,"6":0.363845}
1 0% among other models typesafe/jev-1.13 4/10 runs
+ 4 responses

typesafe/jev-1.13

Prompt 1

{"1":0.29,"2":0.1,"3":0.24,"4":0.21,"5":0.06,"6":0.1}

Prompt 1

{"1":0.28,"2":0.09,"3":0.25,"4":0.22,"5":0.06,"6":0.1}

Prompt 1

{"1":0.29,"2":0.1,"3":0.25,"4":0.22,"5":0.05,"6":0.09}

Prompt 1

{"1":0.27,"2":0.1,"3":0.26,"4":0.21,"5":0.06,"6":0.1}
2 0.2% among other models deepseek/deepseek-v3.2 4/20 runs qwen/qwen3.5-flash-02-23 2/10 runs
+ 6 responses

deepseek/deepseek-v3.2

Prompt 1 · 4 identical runs

2

qwen/qwen3.5-flash-02-23

Prompt 1

2

Prompt 3

2

Author's note

This is not an advice or opinion question. I just find it highly entertaining that "pick a random number" has one of the higest alignments between models out of any of these questions.

User message(s)

Roll a d6
+ 2 more prompts 
Pick a random number between 1 and 6
I need a random number 1-6

106 models

Anthropic

anthropic/claude-haiku-5.5 (10 runs)

4 90%
3 10%
+ 10 responses

anthropic/claude-sonnet-5.5 (10 runs)

4 100%
+ 10 responses

anthropic/claude-opus-5.5 (10 runs)

4 100%
+ 10 responses

anthropic/claude-opus-5 (20 runs)

4 100%
+ 20 responses

anthropic/claude-sonnet-5 (10 runs)

4 100%
+ 10 responses

anthropic/claude-fable-5 (10 runs)

4 100%
+ 10 responses

anthropic/claude-opus-4.8 (10 runs)

4 100%
+ 10 responses

anthropic/claude-opus-4.7 (10 runs)

4 100%
+ 10 responses

anthropic/claude-sonnet-4.6 (5 runs)

4 100%
+ 5 responses

anthropic/claude-opus-4.6 (5 runs)

4 100%
+ 5 responses

anthropic/claude-haiku-4.5 (20 runs)

4 100%
+ 20 responses

anthropic/claude-sonnet-4.5 (5 runs)

4 100%
+ 5 responses

Cloudflare

cloudflare/clef-omni (10 runs)

4 100%
+ 10 responses

cloudflare/clef (10 runs)

4 100%
+ 10 responses

cloudflare/clef-flash (10 runs)

4 100%
+ 10 responses

DeepSeek

deepseek/deepseek-v4.1-flash-1017 (10 runs)

4 100%
+ 10 responses

deepseek/deepseek-v4-flash (20 runs)

4 70%
5 10%
6 10%
+ 20 responses

deepseek/deepseek-v4-pro (20 runs)

4 60%
3 15%
5 25%
+ 20 responses

deepseek/deepseek-v3.2 (20 runs)

4 10%
3 65%
2 20%
+ 20 responses

Google

google/gemini-4-argon (10 runs)

4 100%
+ 10 responses

google/gemini-3.5-flash (10 runs)

4 100%
+ 10 responses

google/gemini-3.1-flash-lite (10 runs)

4 100%
+ 10 responses

google/gemma-4-26b-a4b-it (20 runs)

4 100%
+ 20 responses

google/gemma-4-31b-it (10 runs)

4 100%
+ 10 responses

google/gemini-3-flash-preview (5 runs)

4 100%
+ 5 responses

google/gemini-2.5-flash-lite (20 runs)

4 75%
3 15%
5 10%
+ 20 responses

google/gemini-2.5-flash (15 runs)

4 67%
3 33%
+ 15 responses

IBM

ibm-granite/granite-4.1-8b (10 runs)

4 100%
+ 10 responses

Inception

inception/mercury-decide (10 runs)

3 100%
+ 10 responses

inception/mercury-2.5 (9 runs)

4 67%
3 11%
5 22%
+ 9 responses

inception/mercury-2 (10 runs)

4 90%
3 10%
+ 10 responses

Jared Palmer

jaredpalmer/kev-4b (10 runs)

4 70%
3 30%
+ 10 responses

Liquid AI

liquid/lfm-2.5-24b-a2b (10 runs)

4 100%
+ 10 responses

liquid/d1 (10 runs)

4 40%
3 60%
+ 10 responses

liquid/lfm-2.5-2.6b:free (8 runs)

4 13%
3 88%
+ 8 responses

Meta

meta/muse-spark-open (10 runs)

4 100%
+ 10 responses

meta/muse-wonder (10 runs)

4 90%
5 10%
+ 10 responses

muse-spark-1.1 (20 runs)

4 95%
+ 20 responses

Microsoft

microsoft/microsoft-decision-1 (10 runs)

4 100%
+ 10 responses

MiniMax

minimax/minimax-m3.1-flash (10 runs)

4 90%
5 10%
+ 10 responses

minimax/minimax-m3 (15 runs)

4 87%
+ 15 responses

minimax/minimax-m2.7 (15 runs)

4 93%
+ 15 responses

minimax/minimax-m2.5 (5 runs)

4 100%
+ 5 responses

minimax/minimax-m2.1 (5 runs)

4 100%
+ 5 responses

Mistral

mistralai/mistral-large-4-0 (10 runs)

4 70%
3 30%
+ 10 responses

mistralai/mistral-small-2603 (20 runs)

4 45%
3 55%
+ 20 responses

mistralai/mistral-small-3.2-24b-instruct (20 runs)

4 40%
3 60%
+ 20 responses

mistralai/mistral-nemo (20 runs)

4 20%
3 80%
+ 20 responses

MoonshotAI

moonshotai/kimi-k3.1-pro (10 runs)

4 100%
+ 10 responses

moonshotai/kimi-k3 (20 runs)

4 100%
+ 20 responses

moonshotai/kimi-k2.7-code (15 runs)

4 87%
+ 15 responses

moonshotai/kimi-k2.6 (20 runs)

4 60%
3 40%
+ 20 responses

moonshotai/kimi-k2.5 (10 runs)

4 100%
+ 10 responses

NVIDIA

nvidia/nemotron-3.5-super-vl (10 runs)

4 100%
+ 10 responses

nvidia/nemotron-3-ultra-550b-a55b (10 runs)

4 100%
+ 10 responses

OpenAI

openai/gpt-6-luna-decisions (10 runs)

4 60%
6 40%
+ 10 responses

openai/gpt-6.1-sol (10 runs)

4 100%
+ 10 responses

openai/gpt-6-luna (10 runs)

4 100%
+ 10 responses

openai/gpt-6-sol (10 runs)

4 100%
+ 10 responses

openai/gpt-5.6-luna (20 runs)

4 100%
+ 20 responses

openai/gpt-5.6-sol (20 runs)

4 100%
+ 20 responses

openai/gpt-5.6-terra (20 runs)

4 100%
+ 20 responses

openai/gpt-5.5 (10 runs)

4 100%
+ 10 responses

openai/gpt-5.4-mini (10 runs)

4 100%
+ 10 responses

openai/gpt-5.4-nano (10 runs)

4 100%
+ 10 responses

openai/gpt-5.4 (10 runs)

4 100%
+ 10 responses

openai/gpt-5.3-chat (5 runs)

4 100%
+ 5 responses

openai/gpt-oss-120b (10 runs)

4 100%
+ 10 responses

openai/gpt-4.1-mini (20 runs)

4 100%
+ 20 responses

openai/gpt-4o-mini (5 runs)

4 100%
+ 5 responses

openai/gpt-5.2 (5 runs)

4 100%
+ 5 responses

Qwen

qwen/qwen3.7-plus (15 runs)

4 80%
3 13%
+ 15 responses

qwen/qwen3.7-max (10 runs)

4 100%
+ 10 responses

qwen/qwen3.6-27b (15 runs)

4 73%
3 27%
+ 15 responses

qwen/qwen3.6-flash (10 runs)

4 100%
+ 10 responses

qwen/qwen3.6-max-preview (10 runs)

4 100%
+ 10 responses

qwen/qwen3.6-plus (10 runs)

4 100%
+ 10 responses

qwen/qwen3.5-122b-a10b (10 runs)

4 80%
3 10%
5 10%
+ 10 responses

qwen/qwen3.5-flash-02-23 (10 runs)

4 50%
5 20%
6 10%
2 20%
+ 10 responses

qwen/qwen3-235b-a22b-2507 (5 runs)

4 100%
+ 5 responses

Sakana

sakana/fugu-ultra (10 runs)

4 100%
+ 10 responses

StepFun

stepfun/step-3.7-flash (20 runs)

4 45%
3 10%
5 45%
+ 20 responses

Tencent

tencent/hy3:free (20 runs)

4 85%
3 10%
+ 20 responses

ThinkingMachines

thinkingmachines/inkling (20 runs)

4 100%
+ 20 responses

TypeSafe

typesafe/jev-1.14 (10 runs)

4 90%
3 10%
+ 10 responses

typesafe/jev-1.13 (10 runs)

3 60%
1 40%
+ 10 responses

Upstage

upstage/solar-decide (10 runs)

6 100%
+ 10 responses

upstage/solar-mini4 (10 runs)

4 60%
3 30%
5 10%
+ 10 responses

upstage/solar-pro4 (10 runs)

4 50%
3 30%
5 20%
+ 10 responses

upstage/solar-pro-3 (10 runs)

4 20%
3 40%
5 20%
6 20%
+ 10 responses

xAI

x-ai/grok-4.7 (10 runs)

4 100%
+ 10 responses

x-ai/grok-4.5 (20 runs)

4 90%
+ 20 responses

x-ai/grok-4.3 (20 runs)

4 70%
3 25%
+ 20 responses

x-ai/grok-4.20 (20 runs)

4 95%
+ 20 responses

x-ai/grok-4-fast (10 runs)

4 80%
3 20%
+ 10 responses

x-ai/grok-4.1-fast (15 runs)

4 27%
3 67%
+ 15 responses

Xiaomi

xiaomi/mimo-v2.5 (20 runs)

4 95%
+ 20 responses

xiaomi/mimo-v2.5-pro (20 runs)

4 85%
5 10%
+ 20 responses

xiaomi/mimo-v2-omni (10 runs)

4 100%
+ 10 responses

xiaomi/mimo-v2-pro (10 runs)

4 100%
+ 10 responses

Z.ai

z-ai/glm-5.2 (15 runs)

4 93%
+ 15 responses

z-ai/glm-5.1 (15 runs)

4 73%
3 27%
+ 15 responses

z-ai/glm-5-turbo (10 runs)

4 100%
+ 10 responses

z-ai/glm-5 (10 runs)

4 100%
+ 10 responses

z-ai/glm-4.7-flash (20 runs)

4 65%
3 30%
+ 20 responses

z-ai/glm-4.7 (20 runs)

4 95%
+ 20 responses