DeepSeek V3.1 Terminus
VS
Gemini 3 Flash Preview
DeepSeek V3.1 Terminus vs Gemini 3 Flash Preview: 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.
Gemini 3 Flash Preview wins 18 of 22 task areas we tested; DeepSeek V3.1 Terminus takes 4. DeepSeek V3.1 Terminus costs 2.9× less per token ($1.22 vs $3.5 per 1M).
DeepSeek V3.1 Terminus costs 2.9× less per token ($1.22 vs $3.5 per 1M).
Task by task
| Task area | DeepSeek V3.1 Terminus | Gemini 3 Flash Preview | Winner |
|---|---|---|---|
| RAG, Safety & Grounding |
#85
/ 113
Excellent
|
#13
/ 113
Excellent
|
Gemini 3 Flash Preview |
| Coding |
#92
/ 115
Usable
|
#22
/ 115
Strong
|
Gemini 3 Flash Preview |
| AI Strategy |
#101
/ 126
Usable
|
#38
/ 126
Strong
|
Gemini 3 Flash Preview |
| Sales |
#97
/ 110
Needs editing
|
#35
/ 110
Usable
|
Gemini 3 Flash Preview |
| Content & Brand |
#100
/ 124
Usable
|
#39
/ 124
Strong
|
Gemini 3 Flash Preview |
| Chef / Home Cooking |
#95
/ 126
Usable
|
#39
/ 126
Usable
|
Gemini 3 Flash Preview |
| Legal & HR |
#95
/ 110
Strong
|
#47
/ 110
Excellent
|
Gemini 3 Flash Preview |
| Presentations & Decks |
#91
/ 110
Strong
|
#52
/ 110
Excellent
|
Gemini 3 Flash Preview |
| Creative & Comedy | #89 / 110 | #51 / 110 | Gemini 3 Flash Preview |
| Customer Support |
#69
/ 113
Usable
|
#34
/ 113
Strong
|
Gemini 3 Flash Preview |
| Translation & Localization |
#101
/ 110
Strong
|
#67
/ 110
Excellent
|
Gemini 3 Flash Preview |
| Product & Project Management |
#36
/ 110
Strong
|
#69
/ 110
Strong
|
DeepSeek V3.1 Terminus |
| Frontend & Landing Pages |
#33
/ 109
Needs editing
|
#1
/ 109
Usable
|
Gemini 3 Flash Preview |
| Structured Output |
#63
/ 113
Excellent
|
#34
/ 113
Excellent
|
Gemini 3 Flash Preview |
| Summarization & Meeting Notes |
#32
/ 110
Excellent
|
#61
/ 110
Excellent
|
DeepSeek V3.1 Terminus |
| Research & Competitive Analysis |
#46
/ 110
Strong
|
#71
/ 110
Usable
|
DeepSeek V3.1 Terminus |
| Training & Education |
#31
/ 110
Excellent
|
#11
/ 110
Excellent
|
Gemini 3 Flash Preview |
| Investor & Pitch |
#64
/ 66
Needs editing
|
#47
/ 66
Usable
|
Gemini 3 Flash Preview |
| Executive Assistant |
#77
/ 112
Usable
|
#92
/ 112
Usable
|
DeepSeek V3.1 Terminus |
| Data & Analytics |
#56
/ 110
Excellent
|
#47
/ 110
Excellent
|
Gemini 3 Flash Preview |
| Landing Pages |
#44
/ 72
Strong
|
#35
/ 72
Strong
|
Gemini 3 Flash Preview |
| Knowledge & Docs |
#50
/ 110
Usable
|
#49
/ 110
Usable
|
Gemini 3 Flash Preview |
Rank = position among every model config we tested in that task area (lower is better). Sorted by biggest gap first.
Frequently asked
Is DeepSeek V3.1 Terminus better than Gemini 3 Flash Preview?
Across 22 task areas we benchmarked, Gemini 3 Flash Preview ranks higher in 18 and DeepSeek V3.1 Terminus in 4.
Which is cheaper, DeepSeek V3.1 Terminus or Gemini 3 Flash Preview?
DeepSeek V3.1 Terminus costs 2.9× less per token ($1.22 vs $3.5 per 1M).
What is DeepSeek V3.1 Terminus better at?
DeepSeek V3.1 Terminus out-ranks Gemini 3 Flash Preview at Product & Project Management, Summarization & Meeting Notes, Research & Competitive Analysis.
What is Gemini 3 Flash Preview better at?
Gemini 3 Flash Preview out-ranks DeepSeek V3.1 Terminus at RAG, Safety & Grounding, Coding, AI Strategy.
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