Claude Opus 4.7 vs DeepSeek V4 Pro
Full profiles: Claude Opus 4.7 (Anthropic) and DeepSeek V4 Pro (DeepSeek): every benchmark, price and rank.
Updated · Data from 11 sources
As of 16 Nov 2026, Claude Opus 4.7 is ahead of DeepSeek V4 Pro on quality (63.1 vs 56.0 on the BenchLeader Index), but DeepSeek V4 Pro is 10× cheaper at $0.990 vs $10.00 per 1M tokens blended. Claude Opus 4.7 leads in every category we measure both in.
Key facts (as of 16 Nov 2026)
- Claude Opus 4.7 scores 63.1 on the BenchLeader Index and DeepSeek V4 Pro scores 56.0 (BenchLeader, 16 Nov 2026).
- Claude Opus 4.7 costs $10.00 per 1M tokens blended (list price, 16 Nov 2026).
- DeepSeek V4 Pro costs $0.990 per 1M tokens blended (list price, 16 Nov 2026).
- Of the 12 benchmarks both models have, Claude Opus 4.7 leads 11 (BenchLeader, 16 Nov 2026).
- On Arena WebDev, Claude Opus 4.7 is rated 1555 and DeepSeek V4 Pro 1581 (LMArena, as published by 16 Nov 2026).
Which to pick for…
The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.
| Use case | Pick | Why | Confidence |
|---|---|---|---|
| Coding | Claude Opus 4.7 | 61 vs 52 category score; WeirdML 76.4% vs 46.5% | High |
| Maths | Claude Opus 4.7 | 67 vs 65 category score; Arena Maths 1490 vs 1465 | Medium |
| Reasoning | Claude Opus 4.7 | 64 vs 56 category score; Arena Hard Prompts 1519 vs 1482 | High |
| Chat quality (human preference) | Claude Opus 4.7 | 67 vs 65 on Arena votes | Medium |
| Lowest cost | DeepSeek V4 Pro | $1.98 vs $10.00 per 1M blended; agentic coding session $0.095 vs $0.400 | – |
| Fastest first answer | Claude Opus 4.7 | 1.0 s (Claude Opus 4.7) vs 42.8 s (DeepSeek V4 Pro (high)) to the first answer token | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- Claude Opus 4.7 leads on quality: 63.1 vs 56.0.
- Claude Opus 4.7 is stronger in coding, human preference, maths, reasoning.
- DeepSeek V4 Pro (high) is 10× cheaper ($0.990 vs $10.00 per 1M blended).
- DeepSeek V4 Pro (high) streams 1.7× faster (80 vs 48 tokens per second).
At a glance
| Spec | Claude Opus 4.7 | DeepSeek V4 Pro |
|---|---|---|
| Developer | Anthropic | DeepSeek |
| Released | 16 Apr 2026 | 13 Aug 2026 |
| Context window (tokens) | 1,000,000 | 1,048,576 |
| Compared at | Claude Opus 4.7 | DeepSeek V4 Pro (high) |
| Open weights | No | Yes |
| Price in / out / cached input per 1M | $5.00 / $25.00 / $0.500 | $1.32 / $3.96 / $0.022 |
Category by category
Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.
| Category | Claude Opus 4.7 | DeepSeek V4 Pro (high) | Leader |
|---|---|---|---|
| Reasoning | 63.6 | 55.6 | Claude Opus 4.7 |
| Coding | 60.9 | 51.5 | Claude Opus 4.7 |
| Agents & tools | 67.9 | – | not comparable |
| Maths | 66.7 | 64.5 | Claude Opus 4.7 |
| Knowledge | – | 41.2 | not comparable |
| Human preference | 67.3 | 65.0 | Claude Opus 4.7 |
| Multimodal | 67.1 | – | not comparable |
| Long context | – | 64.3 | not comparable |
On one chart
Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.
Not drawn, measured for fewer than two of these models: Agents & tools, Knowledge, Instruction following, Long context, Multimodal.
Where sources disagree
Across the 12 benchmarks both have, Claude Opus 4.7 leads 11, DeepSeek V4 Pro leads 1. Every overall index that rates both puts Claude Opus 4.7 ahead.
| Overall index | Claude Opus 4.7 | DeepSeek V4 Pro (high) | Leader |
|---|---|---|---|
| BenchLeader Index | 63.1 | 56.0 | Claude Opus 4.7 |
| Arena Text | 1494 | 1464 | Claude Opus 4.7 |
Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).
Price, speed and latency
List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.
| Measure | Claude Opus 4.7 | DeepSeek V4 Pro (high) |
|---|---|---|
| Blended price per 1M tokens | $10.00 | $1.98 |
| Input / output per 1M | $5.00 / $25.00 | $1.32 / $3.96 |
| Cached input per 1M | $0.500 | $0.022 |
| Price per task: chat reply | $0.0095 | $0.0017 |
| Price per task: summarize a 30-page report | $0.075 | $0.018 |
| Price per task: code edit | $0.068 | $0.014 |
| Price per task: agentic coding session | $0.400 | $0.095 |
| Price per task: structured extraction | $0.015 | $0.0034 |
| Agentic coding session, input mostly cached | $0.198 | $0.037 |
| Cost per benchmark task (AA Index) | – | $0.849 |
| Output tokens per task (AA Index) | – | 70k |
| Of which reasoning | – | 45k |
| Time per task (AA Index) | – | 15 min |
| Output speed | 48 tok/s | 80 tok/s |
| Time to first answer | 1.0 s | 42.8 s |
Each model is compared at its best-ranked reasoning-effort setting. Claude Opus 4.7 scores 63.1 and takes 1 s to its first answer token; Claude Opus 4.7 (high) scores 60.0 in 1 s, Claude Opus 4.7 (max) scores 59.7 in 15 s. DeepSeek V4 Pro (high) scores 56.0 and takes 43 s to its first answer token; DeepSeek V4 Pro (max) scores 54.7 in 47 s, DeepSeek V4 Pro scores 53.0 in 1 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.
Every benchmark
152 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.
| Metric | Publisher | Claude Opus 4.7 | DeepSeek V4 Pro (high) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 63.1 | 56.0 |
| Agents & tools score | BenchLeader | 67.9 | – |
| Coding score | BenchLeader | 60.9 | 51.5 |
| Human preference score | BenchLeader | 67.3 | 65.0 |
| Maths score | BenchLeader | 66.7 | 64.5 |
| Multimodal score | BenchLeader | 67.1 | – |
| Reasoning score | BenchLeader | 63.6 | 55.6 |
| Knowledge score | BenchLeader | – | 41.2 |
| Long context score | BenchLeader | – | 64.3 |
| Blended price $/M | BenchLeader | $10.00 | $1.98 |
| Output speed | BenchLeader | 48 tok/s | 80 tok/s |
| Time to first answer | BenchLeader | 1.0 s | 42.8 s |
| Context window | BenchLeader | 1,000,000 tokens | 1,048,576 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 86.4%† | 90.9% |
| Humanity's Last Exam | Scale AI / CAIS | 36.2% | – |
| SimpleBench | SimpleBench | 61.7% | – |
| Arena Hard Prompts | Arena (LMArena) | 1519 | 1482 |
| LiveBench 1.0 Reasoning | LiveBench | 87.2%† | 85.8%† |
| GPQA Diamond (AA) | Artificial Analysis | 91.4%† | 92.8%† |
| Humanity's Last Exam (AA) | Artificial Analysis | 42.3%† | 41.5% |
| HLE-Diamond (AA) | Artificial Analysis | – | 20.5% |
| GPQA Diamond (Vals) | Vals AI | 90.2%† | 92.4%† |
| Kagi LLM Benchmark | Kagi LLM Benchmark | 73.3% | 53.5%† |
| ARC-AGI-1 | ARC Prize | – | 87.2% |
| ARC-AGI-2 | ARC Prize | – | 59.7% |
| ARC-AGI-3 | ARC Prize | 0.2%† | – |
| CritPt | Artificial Analysis | 12.0%† | 18.6% |
| CritPt v2 hard subset (AA) | Artificial Analysis | – | 28.5% |
| HLE-Diamond (Scale) | Scale AI SEAL | – | 13.4%† |
| Arena Expert | Arena (LMArena) | 1534 | 1481 |
| Chess Puzzles | Epoch AI Benchmarking Hub | 41.0%† | 16.0% |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 13.0% | 43.0%† |
| CL-bench Life | Epoch AI Benchmarking Hub | – | 13.5% |
| LMCA | Epoch AI Benchmarking Hub | 52.2%† | 45.5%† |
| DTBench | Epoch AI Benchmarking Hub | 94.7%† | 93.9%† |
| ForecastBench | ForecastBench | 62.8% | 60.7%† |
| Coding | |||
| SWE-bench Verified (Epoch) | Epoch AI Benchmarking Hub | 83.5%† | 77.6%† |
| SciCode | SciCode | 54.5%† | 46.4% |
Show all 152 benchmarks
| Metric | Publisher | Claude Opus 4.7 | DeepSeek V4 Pro (high) |
|---|---|---|---|
| Coding | |||
| WeirdML | WeirdML | 76.4% | 46.5% |
| FrontierCode | Cognition | 38.5% | 28.6% |
| GSO-Bench | GSO-Bench | 44.1% | – |
| Arena Coding | Arena (LMArena) | 1547 | 1503 |
| Arena WebDev | Arena (LMArena) | 1555 | 1581 |
| LiveBench 1.0 Coding | LiveBench | 82.1%† | 77.2%† |
| SciCode (AA) | Artificial Analysis | – | 51.7% |
| LiveCodeBench | Vals AI | 85.1%† | 87.5%† |
| IOI | Vals AI | – | 51.6%† |
| SWE-bench (Vals) | Vals AI | 82.0%† | 96.4%† |
| Code Migration | Vals AI | 43.9%† | 41.5%† |
| Code Migration v1.1 | Vals AI | – | 52.0%† |
| ProgramBench | Vals AI | 0.0%† | 0.0%† |
| Vibe Code Bench 1-100 | Vals AI | – | 17.5%† |
| Vibe Code Bench v1.1 | Vals AI | 71.0%† | 82.3%† |
| SWE Atlas v1.0: Codebase QnA | Scale AI SEAL | 40.3% | 27.1%† |
| SWE Atlas v1.0: Refactoring | Scale AI SEAL | 48.6% | – |
| SWE Atlas v1.0: Test Writing | Scale AI SEAL | 38.5% | 27.1%† |
| MirrorCode | Epoch AI Benchmarking Hub | 31.1%† | 33.3%† |
| Surface Evolver Bench | Epoch AI Benchmarking Hub | – | 40.0% |
| ALE-Bench | Epoch AI Benchmarking Hub | 1323.0 | 1006.1 |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | – | 6.1% |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | 83.2%† | 78.7%† |
| Startrise Frontend | Startrise | – | 62.7%† |
| GameBench 2 Overall | DuelLab GameBench 2 | – | 41.9† |
| GameBench 2 by reasoning setting | DuelLab GameBench 2 | – | 45.1 |
| Design Arena Website | Design Arena | – | 1227† |
| FrontierCode 1.1 (AA) | Artificial Analysis | – | 38.8%† |
| FrontierCode 2.0 (AA) | Artificial Analysis | – | 16.7% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 80.2% | 6.7%† |
| OSWorld-Verified 2.0 | OSWorld | 18.2%† | – |
| APEX-Agents | Mercor | 49.2%† | 47.3%† |
| LiveBench Agentic Coding | LiveBench | 50.7%† | 55.0%† |
| Terminal-Bench Hard | Artificial Analysis | 51.5%† | – |
| τ²-Bench Telecom (AA) | Artificial Analysis | 88.6%† | – |
| Terminal-Bench 2.1 (Vals) | Vals AI | 68.5%† | 50.2%† |
| MCP Atlas | Scale AI SEAL | 79.1%† | – |
| HiL-Bench | Scale AI SEAL | 41.7% | 38.5%† |
| τ³-bench Banking | τ²-bench | 40.2%† | – |
| GDPval-AA v2.1, normalised score | Artificial Analysis | – | 47.5%† |
| GDPval-AA v2.1 | Artificial Analysis | 1361† | 1336 |
| τ³-Banking (AA) | Artificial Analysis | 34.6%† | 39.6%† |
| ITBench SRE (AA) | Artificial Analysis | 46.7%† | – |
| Analyst Agent (AA) | Artificial Analysis | 43.8%† | – |
| CyberBench | Vals AI | – | 72.9%† |
| Finance Agent v2 | Vals AI | 51.5%† | 50.4%† |
| Finance Agent v2.1 | Vals AI | – | 57.1%† |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | – | 10.8%† |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 6.7%† | 7.5%† |
| Legal Research Bench | Vals AI | 38.5%† | 40.9%† |
| SkillsBench | Vals AI | – | 53.8%† |
| SREBench | Vals AI | – | 1.9%† |
| Tax Agent Bench | Vals AI | 23.7%† | 26.8%† |
| Terminal-Bench 2.0 (Vals) | Vals AI | 68.5%† | 56.2%† |
| Terminal-Bench 4.0 (Vals) | Vals AI | – | 14.1%† |
| Terminal-Bench Science | Vals AI | – | 4.3%† |
| Arena Search | Arena (LMArena) | 1210 | – |
| PostTrainBench | Epoch AI Benchmarking Hub | 24.3%† | – |
| ExploitBench | Epoch AI Benchmarking Hub | 26.5% | – |
| GBAEval | Epoch AI Benchmarking Hub | 43.8% | – |
| Vending-Bench 2 | Andon Labs | 10936.8 | 3284.5† |
| AutomationBench | Artificial Analysis | – | 46.2% |
| GDP.pdf (AA) | Artificial Analysis | – | 10.4% |
| EnterpriseOps-Gym | Artificial Analysis | – | 49.6%† |
| AA-Briefcase v1.1 | Artificial Analysis | 1252† | 1148 |
| Design Arena Agentic | Design Arena | – | 927† |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | – | 1.9% |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 70.2%† | 64.6%† |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 31.7%† | 26.8%† |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 86.7%† | 95.6% |
| ProofBench | Vals AI | 54.0%† | 50.0%† |
| LiveBench Mathematics | LiveBench | 92.8%† | 95.1%† |
| AIME (Vals) | Vals AI | 96.3% | – |
| AIME 2026 | MathArena | 95.8%† | 96.7%† |
| HMMT February 2026 | MathArena | 93.9%† | 93.9%† |
| USAMO 2026 | MathArena | – | 60.7%† |
| MathArena Apex | MathArena | 40.6%† | 28.1%† |
| Arena Maths | Arena (LMArena) | 1490 | 1465 |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 51.7%† | 52.9%† |
| LiveBench 1.0 Data Analysis | LiveBench | 78.3%† | 79.2%† |
| AA-Omniscience v1.1 | Artificial Analysis | – | -6.5 |
| MMLU-Pro | Vals AI | 89.9%† | 87.3%† |
| LegalBench | Vals AI | 85.3%† | 82.4%† |
| CorpFin | Vals AI | 66.1%† | 65.4%† |
| TaxEval | Vals AI | 75.3%† | 73.1%† |
| CaseLaw v2 | Vals AI | 68.4% | 59.4%† |
| Excel Modeling Benchmark | Vals AI | 63.7%† | 52.8%† |
| Excel Modeling Benchmark v1.1 | Vals AI | – | 65.1%† |
| MedCode | Vals AI | 54.9%† | 42.5%† |
| MedScribe | Vals AI | 83.0%† | 80.2%† |
| Public Benefits Bench v1.1 | Vals AI | – | 62.9%† |
| Immigration Bench | Vals AI | – | 61.7%† |
| DrugDiscoveryBench | Scale AI SEAL | – | 31.7%† |
| EBR-bench | Epoch AI Benchmarking Hub | 19.1%† | – |
| GDP.pdf | Epoch AI Benchmarking Hub | 21.0%† | – |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | – | 45.1% |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | – | 5.9% |
| Instruction following | |||
| LiveBench Language | LiveBench | 77.9%† | 82.1%† |
| LiveBench Instruction Following | LiveBench | 66.7%† | 67.7%† |
| IFBench | Artificial Analysis | 58.6%† | – |
| Arena Instruction Following | Arena (LMArena) | 1493 | 1461 |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1494 | 1464 |
| EQ-Bench 4 | EQ-Bench | 1293 | 1156† |
| EQ-Bench Creative Writing v3 | EQ-Bench | 1915 | 1553† |
| Arena Creative Writing | Arena (LMArena) | 1482 | 1444 |
| Arena Multi-turn | Arena (LMArena) | 1514 | 1469 |
| Arena Longer Queries | Arena (LMArena) | 1508 | 1474 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1319 | – |
| MMMU-Pro | Artificial Analysis | 78.8%† | – |
| MMMU-Pro (Vals) | Vals AI | 85.5%† | – |
| MortgageTax | Vals AI | 70.3%† | – |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 24.5% | – |
| Furniture Assembly | Epoch AI Benchmarking Hub | 33.3%† | – |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 78.7%† | 79.3% |
| Arena Document | Arena (LMArena) | 1494 | – |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 157.4 | 155.9† |
| LiveBench 1.0 | LiveBench | 76.5%† | 77.4%† |
| AA Intelligence Index v5.0.1 | Artificial Analysis | – | 20.8 |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 40.7† | 36† |
| AA Progress Index v1 | Artificial Analysis | 312.5† | 311.5† |
| Vals Index | Vals AI | – | 53.0† |
| Vals Finance Index | Vals AI | – | 52.9%† |
| Vals RSI Index | Vals AI | 19.1%† | – |
| Vals Multimodal Index | Vals AI | 67.4%† | – |
| Vals Router Index | Vals AI | – | 53.0%† |
| Vals Legal Index | Vals AI | – | 69.0%† |
| Metadata | |||
| AA Openness Index | Artificial Analysis | – | 44.4 |
Data as of 2026-11-16. Best configuration of each model; every score links to its source on the model pages.
DuelLab GameBench 2, release gb2-20261111-7b598bb13de4 (https://benchmarks.duellab.org/releases/gb2-20261111-7b598bb13de4/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.
What changed
- DeepSeek V4 Pro (high): BenchLeader Index 59.1 → 56.0.
- DeepSeek V4 Pro (high): 17 new results (AA Intelligence Index v5.0.1, Humanity's Last Exam (AA), HLE-Diamond (AA)…); input price moved from $0.66/M to $1.32/M (100% up); output price moved from $1.98/M to $3.96/M (100% up).
- DeepSeek V4 Pro (high): BenchLeader Index 57.5 → 59.1.
- Claude Opus 4.7: 1 score revision.
- DeepSeek V4 Pro (high): 1 score revision.
- DeepSeek V4 Pro (high): 1 score revision.
- Claude Opus 4.7: 1 score revision.
- Claude Opus 4.7: 1 new result (Harvey's Legal Agent Benchmark v1).
- DeepSeek V4 Pro (high): 16 new results (GPQA Diamond, OTIS Mock AIME, SciCode…); input price moved from $0.44/M to $0.66/M (50% up); output price moved from $0.87/M to $1.98/M (128% up).
- Claude Opus 4.7: 1 score revision.
Related comparisons
- Claude Opus 4.7 vs Claude Fable 5.2
- Claude Opus 4.7 vs Gemini 4 Carbon
- Claude Opus 4.7 vs Claude Opus 5.5
- Claude Opus 4.7 vs GPT-6 Astra
- DeepSeek V4 Pro vs Claude Fable 5.2
- DeepSeek V4 Pro vs Claude Opus 5.5
- Claude Opus 4.7 vs Claude Fable 5.1
- Claude Opus 4.7 vs GPT-6.1 Sol
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
Claude Opus 4.7 vs DeepSeek V4 Pro: questions
- Is Claude Opus 4.7 better than DeepSeek V4 Pro?
- Claude Opus 4.7 leads on quality: 63.1 vs 56.0. The BenchLeader Index combines every independent quality benchmark; Claude Opus 4.7 is ahead overall as of 16 Nov 2026, but check the category scores for your use.
- Claude Opus 4.7 vs DeepSeek V4 Pro: which is cheaper, and will that change?
- DeepSeek V4 Pro is cheaper today: $0.990 against $10.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is Claude Opus 4.7 or DeepSeek V4 Pro better for coding?
- Claude Opus 4.7 scores higher in coding (61 vs 52 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1555 for Claude Opus 4.7 against 1581 for DeepSeek V4 Pro.
- Which is faster, Claude Opus 4.7 or DeepSeek V4 Pro?
- DeepSeek V4 Pro streams faster (80 vs 48 output tokens per second, Artificial Analysis 7-day median).
- Which is better for long documents?
- Both take about 1M tokens: Claude Opus 4.7 1,000,000, DeepSeek V4 Pro 1,048,576.
More questions (3)
- Which has lower latency?
- Claude Opus 4.7 answers first: 1.0 s against 42.8 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
- Do independent benchmarks agree on Claude Opus 4.7 vs DeepSeek V4 Pro?
- Across the 12 benchmarks both have, Claude Opus 4.7 leads 11, DeepSeek V4 Pro leads 1. Every overall index that rates both puts Claude Opus 4.7 ahead.
- When was DeepSeek V4 Pro released?
- DeepSeek V4 Pro was released on 13 Aug 2026.