DeepSeek V3.2
VS
GPT-5 Mini
DeepSeek V3.2 vs GPT-5 Mini: which wins at real work?
22 task areas · same graded test runs · rank comparison only, so 0–100 and Elo collections never mix raw scores.
GPT-5 Mini wins 15 of 22 task areas we tested; DeepSeek V3.2 takes 7. DeepSeek V3.2 costs 3.9× less per token ($0.572 vs $2.25 per 1M).
DeepSeek V3.2 costs 3.9× less per token ($0.572 vs $2.25 per 1M).
Task by task
| Task area | DeepSeek V3.2 | GPT-5 Mini | Winner |
|---|---|---|---|
| Legal & HR |
#90
/ 110
Strong
|
#10
/ 110
Excellent
|
GPT-5 Mini |
| Coding |
#100
/ 115
Usable
|
#37
/ 115
Strong
|
GPT-5 Mini |
| Summarization & Meeting Notes |
#30
/ 110
Excellent
|
#81
/ 110
Excellent
|
DeepSeek V3.2 |
| Structured Output |
#67
/ 113
Strong
|
#17
/ 113
Excellent
|
GPT-5 Mini |
| Training & Education |
#38
/ 110
Excellent
|
#81
/ 110
Strong
|
DeepSeek V3.2 |
| Data & Analytics |
#69
/ 110
Excellent
|
#27
/ 110
Excellent
|
GPT-5 Mini |
| Presentations & Decks |
#81
/ 110
Strong
|
#39
/ 110
Excellent
|
GPT-5 Mini |
| Translation & Localization |
#89
/ 110
Strong
|
#52
/ 110
Excellent
|
GPT-5 Mini |
| Chef / Home Cooking |
#78
/ 126
Usable
|
#45
/ 126
Usable
|
GPT-5 Mini |
| Executive Assistant |
#76
/ 112
Usable
|
#108
/ 112
Usable
|
DeepSeek V3.2 |
| Research & Competitive Analysis |
#53
/ 110
Usable
|
#21
/ 110
Excellent
|
GPT-5 Mini |
| Customer Support |
#74
/ 113
Usable
|
#104
/ 113
Usable
|
DeepSeek V3.2 |
| Content & Brand |
#95
/ 124
Usable
|
#69
/ 124
Strong
|
GPT-5 Mini |
| Product & Project Management |
#21
/ 110
Excellent
|
#1
/ 110
Excellent
|
GPT-5 Mini |
| Creative & Comedy | #87 / 110 | #70 / 110 | GPT-5 Mini |
| Knowledge & Docs |
#58
/ 110
Usable
|
#41
/ 110
Strong
|
GPT-5 Mini |
| Frontend & Landing Pages |
#25
/ 109
Needs editing
|
#12
/ 109
Needs editing
|
GPT-5 Mini |
| Sales |
#91
/ 110
Usable
|
#102
/ 110
Needs editing
|
DeepSeek V3.2 |
| RAG, Safety & Grounding |
#83
/ 113
Excellent
|
#78
/ 113
Excellent
|
GPT-5 Mini |
| Investor & Pitch |
#46
/ 66
Usable
|
#50
/ 66
Usable
|
DeepSeek V3.2 |
| AI Strategy |
#102
/ 126
Usable
|
#100
/ 126
Usable
|
GPT-5 Mini |
| Landing Pages |
#48
/ 72
Usable
|
#50
/ 72
Usable
|
DeepSeek V3.2 |
Rank = position among every model config we tested in that task area (lower is better). Sorted by biggest gap first.
Frequently asked
Is DeepSeek V3.2 better than GPT-5 Mini?
Across 22 task areas we benchmarked, GPT-5 Mini ranks higher in 15 and DeepSeek V3.2 in 7.
Which is cheaper, DeepSeek V3.2 or GPT-5 Mini?
DeepSeek V3.2 costs 3.9× less per token ($0.572 vs $2.25 per 1M).
What is DeepSeek V3.2 better at?
DeepSeek V3.2 out-ranks GPT-5 Mini at Summarization & Meeting Notes, Training & Education, Executive Assistant.
What is GPT-5 Mini better at?
GPT-5 Mini out-ranks DeepSeek V3.2 at Legal & HR, Coding, Structured Output.
More comparisons
This page is Spring Prompt, running in public
We just did this for every model. Do it for your prompt.
The rankings above come from running real tasks through real models and scoring every output. Spring Prompt is that same engine — pointed at your prompt, your test cases, and your definition of good.
- Generate test cases from your prompt — no eval set required to start.
- Compare models side by side with quality, cost and latency in one matrix.
- Optimise the winner until the scores say it's ready to ship.
Prompt × model results
12 test cases · 3 evals