Claude Opus 4.8
VS
DeepSeek V3.1 Terminus
Claude Opus 4.8 vs DeepSeek V3.1 Terminus: 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.
Claude Opus 4.8 wins 21 of 22 task areas we tested; DeepSeek V3.1 Terminus takes 1. DeepSeek V3.1 Terminus costs 24.6× less per token ($1.22 vs $30 per 1M).
DeepSeek V3.1 Terminus costs 24.6× less per token ($1.22 vs $30 per 1M).
Task by task
| Task area | Claude Opus 4.8 | DeepSeek V3.1 Terminus | Winner |
|---|---|---|---|
| AI Strategy |
#4
/ 126
Strong
|
#101
/ 126
Usable
|
Claude Opus 4.8 |
| Sales |
#1
/ 110
Strong
|
#97
/ 110
Needs editing
|
Claude Opus 4.8 |
| Content & Brand |
#8
/ 124
Strong
|
#100
/ 124
Usable
|
Claude Opus 4.8 |
| Legal & HR |
#5
/ 110
Excellent
|
#95
/ 110
Strong
|
Claude Opus 4.8 |
| Coding |
#3
/ 115
Excellent
|
#92
/ 115
Usable
|
Claude Opus 4.8 |
| Chef / Home Cooking |
#9
/ 126
Strong
|
#95
/ 126
Usable
|
Claude Opus 4.8 |
| RAG, Safety & Grounding |
#11
/ 113
Excellent
|
#85
/ 113
Excellent
|
Claude Opus 4.8 |
| Creative & Comedy | #17 / 110 | #89 / 110 | Claude Opus 4.8 |
| Executive Assistant |
#9
/ 112
Strong
|
#77
/ 112
Usable
|
Claude Opus 4.8 |
| Translation & Localization |
#35
/ 110
Excellent
|
#101
/ 110
Strong
|
Claude Opus 4.8 |
| Customer Support |
#8
/ 113
Strong
|
#69
/ 113
Usable
|
Claude Opus 4.8 |
| Presentations & Decks |
#32
/ 110
Excellent
|
#91
/ 110
Strong
|
Claude Opus 4.8 |
| Data & Analytics |
#2
/ 110
Excellent
|
#56
/ 110
Excellent
|
Claude Opus 4.8 |
| Knowledge & Docs |
#2
/ 110
Excellent
|
#50
/ 110
Usable
|
Claude Opus 4.8 |
| Research & Competitive Analysis |
#5
/ 110
Excellent
|
#46
/ 110
Strong
|
Claude Opus 4.8 |
| Investor & Pitch |
#27
/ 66
Strong
|
#64
/ 66
Needs editing
|
Claude Opus 4.8 |
| Product & Project Management |
#2
/ 110
Excellent
|
#36
/ 110
Strong
|
Claude Opus 4.8 |
| Training & Education |
#2
/ 110
Excellent
|
#31
/ 110
Excellent
|
Claude Opus 4.8 |
| Landing Pages |
#18
/ 72
Strong
|
#44
/ 72
Strong
|
Claude Opus 4.8 |
| Frontend & Landing Pages |
#21
/ 109
Needs editing
|
#33
/ 109
Needs editing
|
Claude Opus 4.8 |
| Summarization & Meeting Notes |
#35
/ 110
Excellent
|
#32
/ 110
Excellent
|
DeepSeek V3.1 Terminus |
| Structured Output |
#61
/ 113
Excellent
|
#63
/ 113
Excellent
|
Claude Opus 4.8 |
Rank = position among every model config we tested in that task area (lower is better). Sorted by biggest gap first.
Same task, both models — judged
Both models answered the same test case; an independent judge graded each. Quotes are the judge's actual rationale.
Legal & HR
Termination clause with two conditions (Plain-English Explainer)
“The model perfectly translated the clause into plain English, preserving both the 30-day written notice requirement and the active Service Period exception without changing the legal meaning. It also correctly flagged that 'Service Period' is a defined term that needs to be checked. It did not fabricate any authority. The only minor deduction is for lacking an explicit 'not legal advice' disclaimer, though the tone was appropriately informational and objective.”
“The model commits a severe failure by changing the legal mechanics of the clause. It incorrectly asserts that notice cannot be given during an active Service Period, whereas the clause only dictates that the termination cannot take effect during that time. It also fabricates facts by assuming the Service Period is tied to a 'paid for' billing cycle. Finally, it provides no legal caveats.”
Frequently asked
Is Claude Opus 4.8 better than DeepSeek V3.1 Terminus?
Across 22 task areas we benchmarked, Claude Opus 4.8 ranks higher in 21 and DeepSeek V3.1 Terminus in 1.
Which is cheaper, Claude Opus 4.8 or DeepSeek V3.1 Terminus?
DeepSeek V3.1 Terminus costs 24.6× less per token ($1.22 vs $30 per 1M).
What is Claude Opus 4.8 better at?
Claude Opus 4.8 out-ranks DeepSeek V3.1 Terminus at AI Strategy, Sales, Content & Brand.
What is DeepSeek V3.1 Terminus better at?
DeepSeek V3.1 Terminus out-ranks Claude Opus 4.8 at Summarization & Meeting Notes.
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