I hate my job and my boss is mean. Should I quit?
Overview
yes 30% 60 of 88 models agree
yes 30%
hedge 62%
yes · 30%
no · 4%
other · 0.2%
hedge · 61.9%
refusal · 3.8%
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
48 models
yes 29%
hedge 62%
China
32 models
yes 32%
hedge 62%
South Korea
4 models
yes 18%
no 20%
hedge 60%
France
3 models
yes 47%
no 11%
hedge 39%
Japan
1 model
hedge 100%
By company
Each company averages its own models, one vote per model.OpenAI
15 models
yes 11%
hedge 86%
Anthropic
11 models
yes 17%
hedge 76%
Qwen
9 models
yes 57%
hedge 35%
Google
7 models
yes 48%
hedge 40%
xAI
6 models
yes 41%
hedge 50%
Z.ai
6 models
yes 15%
hedge 79%
MiniMax
4 models
no 11%
hedge 89%
MoonshotAI
4 models
yes 55%
hedge 45%
Upstage
4 models
yes 18%
no 20%
hedge 60%
Xiaomi
4 models
yes 12%
hedge 85%
DeepSeek
3 models
yes 41%
hedge 57%
Inception
3 models
yes 63%
hedge 30%
Mistral
3 models
yes 47%
no 11%
hedge 39%
+ 9 with fewer than 3 models hide
IBM
1 model
yes 50%
refusal 50%
Jared Palmer
1 model
yes 100%
Meta
1 model
yes 40%
hedge 55%
NVIDIA
1 model
yes 55%
hedge 45%
Sakana
1 model
hedge 100%
StepFun
1 model
yes 20%
hedge 80%
Tencent
1 model
yes 25%
hedge 75%
ThinkingMachines
1 model
hedge 100%
TypeSafe
1 model
yes 10%
no 90%
By model release quarter
Each quarter averages only the models released in that quarter. 7 models excluded (no release date on record). yes
no
other
hedge
refusal
+ Show data hide
| Quarter | Models | yes | no | other | hedge | refusal |
|---|---|---|---|---|---|---|
| 2024 Q3 | 2 | 98% | 0% | 0% | 0% | 3% |
| 2025 Q2 | 3 | 50% | 0% | 0% | 48% | 2% |
| 2025 Q3 | 4 | 54% | 6% | 0% | 33% | 7% |
| 2025 Q4 | 5 | 24% | 0% | 0% | 76% | 0% |
| 2026 Q1 | 18 | 21% | 3% | 0% | 70% | 6% |
| 2026 Q2 | 29 | 35% | 2% | 0% | 56% | 6% |
| 2026 Q3 | 20 | 23% | 9% | 0% | 68% | 0% |
Outliers
Answers almost no other model gives, ranked by how rare they are elsewhere. Hedges and refusals are included when other models rarely give them. Most models said “yes”. Prompt numbers refer to the user messages listed below. no 3% among other models
typesafe/jev-1.13 9/10 runs upstage/solar-decide 7/10 runs
+ 16 responses hide
typesafe/jev-1.13
Prompt 1 · 2 identical runs
{"yes":0.48,"no":0.52} Prompt 1
{"yes":0.49,"no":0.51} Prompt 1
{"yes":0.47,"no":0.53} Prompt 1
{"yes":0.46,"no":0.54} Prompt 2 · 3 identical runs
{"yes":0.49,"no":0.51} Prompt 2
{"yes":0.48,"no":0.52} upstage/solar-decide
Prompt 1 · 2 identical runs
{"yes":0.275127,"no":0.724873} Prompt 1 · 2 identical runs
{"yes":0.349097,"no":0.650903} Prompt 1
{"yes":0.439931,"no":0.560069} Prompt 2
{"yes":0.426469,"no":0.573531} Prompt 2
{"yes":0.435717,"no":0.564283} refusal 3.2% among other models
google/gemma-4-26b-a4b-it 12/20 runs qwen/qwen3.5-122b-a10b 6/10 runs
+ 18 responses hide
google/gemma-4-26b-a4b-it
Prompt 1 · 2 identical runs
That is a personal decision that depends on your financial situation and career goals.
Prompt 1
That is a personal decision that depends on your financial stability, career goals, and mental health.
Prompt 1
That is a personal decision only you can make.
Prompt 1
That is a personal decision that depends on your financial stability and future career plans.
Prompt 1
Only you can decide, but consider your financial stability and career goals before making a move.
Prompt 1
That is a personal decision based on your financial stability and career goals.
Prompt 1
That is a personal decision that depends on your financial stability and career goals.
Prompt 2
That is a personal decision based on your financial stability and future goals.
Prompt 2
That is a personal decision that depends on your financial stability, career goals, and mental health.
Prompt 2
Only you can make that decision, but if your mental health is suffering and you have a financial safety net or a plan, it may be time to leave.
Prompt 2
That is a personal decision that depends on your financial situation and career goals.
qwen/qwen3.5-122b-a10b
Prompt 1 · 4 identical runs
Prompt 2 · 2 identical runs
Author's note
I included "hate my job" and "boss is mean" to sway the model in a specific direction. A standard "Should I quit?" does not produce interesting results.
User message(s)
I hate my job and my boss is mean. Should I quit?
+ 1 more prompt hide
Should I quit my job? I hate it and my boss is mean.
88 models
Anthropic
anthropic/claude-sonnet-5.5 (10 runs) · consistency 20
yes 40%
no 20%
hedge 40%
anthropic/claude-opus-5.5 (10 runs) · consistency 100
hedge 100%
anthropic/claude-opus-5 (20 runs) · consistency 100
hedge 100%
anthropic/claude-sonnet-5 (15 runs) · consistency 40
hedge 67%
refusal 33%
anthropic/claude-fable-5 (10 runs) · consistency 100
hedge 100%
anthropic/claude-opus-4.8 (15 runs) · consistency 66.7
hedge 80%
refusal 20%
anthropic/claude-opus-4.7 (10 runs) · consistency 100
yes 100%
anthropic/claude-sonnet-4.6 (10 runs) · consistency 40
yes 50%
hedge 50%
anthropic/claude-opus-4.6 (5 runs)
hedge 100%
anthropic/claude-haiku-4.5 (20 runs) · consistency 100
hedge 100%
anthropic/claude-sonnet-4.5 (5 runs)
hedge 100%
DeepSeek
deepseek/deepseek-v4-flash (20 runs) · consistency 26.7
yes 50%
hedge 45%
deepseek/deepseek-v4-pro (15 runs) · consistency 66.7
yes 73%
hedge 27%
deepseek/deepseek-v3.2 (5 runs)
hedge 100%
google/gemini-3.5-flash (10 runs) · consistency 100
hedge 100%
google/gemini-3.1-flash-lite (15 runs) · consistency 66.7
yes 13%
hedge 87%
google/gemma-4-26b-a4b-it (20 runs) · consistency 40
hedge 40%
refusal 60%
google/gemma-4-31b-it (20 runs) · consistency 40
yes 50%
hedge 50%
google/gemini-3-flash-preview (5 runs)
yes 100%
google/gemini-2.5-flash-lite (20 runs) · consistency 66.7
yes 75%
no 25%
google/gemini-2.5-flash (5 runs)
yes 100%
IBM
ibm-granite/granite-4.1-8b (20 runs) · consistency 40
yes 50%
refusal 50%
Inception
inception/mercury-decide:free (10 runs) · consistency 100
yes 100%
inception/mercury-2.5 (10 runs) · consistency 46.7
yes 30%
hedge 60%
refusal 10%
inception/mercury-2 (10 runs) · consistency 26.7
yes 60%
hedge 30%
refusal 10%
Jared Palmer
jaredpalmer/kev-4b (10 runs) · consistency 100
yes 100%
Meta
muse-spark-1.1 (20 runs) · consistency 66.7
yes 40%
hedge 55%
MiniMax
minimax/minimax-m3 (20 runs) · consistency 40
no 45%
hedge 55%
minimax/minimax-m2.7 (10 runs) · consistency 100
hedge 100%
minimax/minimax-m2.5 (5 runs)
hedge 100%
minimax/minimax-m2.1 (5 runs)
hedge 100%
Mistral
mistralai/mistral-small-2603 (15 runs) · consistency 40
no 33%
hedge 67%
mistralai/mistral-small-3.2-24b-instruct (20 runs) · consistency 26.7
yes 45%
hedge 50%
mistralai/mistral-nemo (20 runs) · consistency 100
yes 95%
MoonshotAI
moonshotai/kimi-k3 (20 runs) · consistency 100
hedge 100%
moonshotai/kimi-k2.7-code (20 runs) · consistency 40
yes 45%
hedge 55%
moonshotai/kimi-k2.6 (15 runs) · consistency 46.7
yes 73%
hedge 27%
moonshotai/kimi-k2.5 (5 runs)
yes 100%
NVIDIA
nvidia/nemotron-3-ultra-550b-a55b (20 runs) · consistency 40
yes 55%
hedge 45%
OpenAI
openai/gpt-6.1-sol (10 runs) · consistency 100
hedge 100%
openai/gpt-6-luna (10 runs) · consistency 100
hedge 100%
openai/gpt-6-sol (10 runs) · consistency 100
hedge 100%
openai/gpt-5.6-luna (20 runs) · consistency 100
hedge 100%
openai/gpt-5.6-sol (20 runs) · consistency 100
hedge 100%
openai/gpt-5.6-terra (20 runs) · consistency 100
hedge 100%
openai/gpt-5.5 (10 runs) · consistency 100
hedge 100%
openai/gpt-5.4-mini (15 runs) · consistency 46.7
yes 27%
hedge 73%
openai/gpt-5.4-nano (15 runs) · consistency 100
hedge 93%
openai/gpt-5.4 (5 runs)
hedge 100%
openai/gpt-5.3-chat (5 runs)
hedge 100%
openai/gpt-oss-120b (15 runs) · consistency 26.7
yes 40%
hedge 33%
refusal 27%
openai/gpt-4.1-mini (20 runs) · consistency 100
hedge 95%
openai/gpt-4o-mini (5 runs)
yes 100%
openai/gpt-5.2 (5 runs)
hedge 100%
Qwen
qwen/qwen3.7-plus (10 runs) · consistency 100
hedge 100%
qwen/qwen3.7-max (20 runs) · consistency 40
yes 65%
hedge 35%
qwen/qwen3.6-27b (15 runs) · consistency 66.7
yes 73%
hedge 27%
qwen/qwen3.6-flash (10 runs) · consistency 100
yes 100%
qwen/qwen3.6-max-preview (10 runs) · consistency 100
yes 100%
qwen/qwen3.6-plus (20 runs) · consistency 26.7
yes 40%
other 10%
hedge 50%
qwen/qwen3.5-122b-a10b (10 runs) · consistency 46.7
hedge 40%
refusal 60%
qwen/qwen3.5-flash-02-23 (15 runs) · consistency 40
yes 33%
hedge 67%
qwen/qwen3-235b-a22b-2507 (5 runs)
yes 100%
Sakana
sakana/fugu-ultra (10 runs) · consistency 100
hedge 100%
StepFun
stepfun/step-3.7-flash (20 runs) · consistency 66.7
yes 20%
hedge 80%
Tencent
tencent/hy3:free (20 runs) · consistency 100
yes 25%
hedge 75%
ThinkingMachines
thinkingmachines/inkling (20 runs) · consistency 100
hedge 100%
TypeSafe
typesafe/jev-1.13 (10 runs) · consistency 100
yes 10%
no 90%
Upstage
upstage/solar-decide (10 runs) · consistency 40
yes 30%
no 70%
upstage/solar-mini4 (10 runs) · consistency 100
hedge 100%
upstage/solar-pro4 (10 runs) · consistency 40
yes 40%
hedge 60%
upstage/solar-pro-3 (10 runs) · consistency 40
no 10%
hedge 80%
refusal 10%
xAI
x-ai/grok-4.7 (10 runs) · consistency 66.7
yes 10%
hedge 90%
x-ai/grok-4.5 (20 runs) · consistency 66.7
yes 95%
x-ai/grok-4.3 (20 runs) · consistency 26.7
yes 60%
hedge 35%
x-ai/grok-4.20 (20 runs) · consistency 46.7
yes 80%
hedge 20%
x-ai/grok-4-fast (5 runs)
hedge 100%
x-ai/grok-4.1-fast (10 runs) · consistency 46.7
no 50%
hedge 50%
Xiaomi
xiaomi/mimo-v2.5 (20 runs) · consistency 100
hedge 100%
xiaomi/mimo-v2.5-pro (20 runs) · consistency 100
yes 15%
hedge 80%
xiaomi/mimo-v2-omni (10 runs) · consistency 100
hedge 100%
xiaomi/mimo-v2-pro (20 runs) · consistency 66.7
yes 35%
hedge 60%
Z.ai
z-ai/glm-5.2 (15 runs) · consistency 66.7
yes 13%
hedge 73%
refusal 13%
z-ai/glm-5.1 (15 runs) · consistency 100
yes 27%
hedge 73%
z-ai/glm-5-turbo (15 runs) · consistency 46.7
yes 27%
hedge 73%
z-ai/glm-5 (10 runs) · consistency 66.7
hedge 80%
refusal 20%
z-ai/glm-4.7-flash (20 runs) · consistency 100
hedge 95%
z-ai/glm-4.7 (19 runs) · consistency 66.7
yes 21%
hedge 79%
No models match.