Claude Opus 4.7 vs Grok 4
Full profiles: Claude Opus 4.7 (Anthropic) and Grok 4 (SpaceXAI): every benchmark, price and rank.
Updated · Data from 13 sources
As of 16 Nov 2026, Claude Opus 4.7 is ahead of Grok 4 on quality (63.1 vs 58.1 on the BenchLeader Index), but Grok 4 is 1.7× cheaper at $6.00 vs $10.00 per 1M tokens blended. Claude Opus 4.7 leads in every category we measure both in. Grok 4 streams ~1.5× faster.
Key facts (as of 16 Nov 2026)
- Claude Opus 4.7 scores 63.1 on the BenchLeader Index and Grok 4 scores 58.1 (BenchLeader, 16 Nov 2026).
- Claude Opus 4.7 costs $10.00 per 1M tokens blended (list price, 16 Nov 2026).
- Grok 4 costs $6.00 per 1M tokens blended (list price, 16 Nov 2026).
- Grok 4 streams ~1.5× faster than Claude Opus 4.7 (Artificial Analysis, 7-day median, 16 Nov 2026).
- Of the 18 benchmarks both models have, Claude Opus 4.7 leads 16 (BenchLeader, 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 56 category score; WeirdML 76.4% vs 45.7% | Medium |
| Agents & tool use | Claude Opus 4.7 | 68 vs 59 category score; Terminal-Bench 80.2% vs 27.2% | High |
| Maths | Claude Opus 4.7 | 67 vs 58 category score; Arena Maths 1490 vs 1425 | High |
| Reasoning | Claude Opus 4.7 | 64 vs 60 category score; SimpleBench 61.7% vs 60.5% | Medium |
| Images and other media | Claude Opus 4.7 | 67 vs 52 category score; Arena Vision 1319 vs 1210 | High |
| Chat quality (human preference) | Claude Opus 4.7 | 67 vs 59 on Arena votes | High |
| Lowest cost | Grok 4 | $6.00 vs $10.00 per 1M blended; agentic coding session $0.240 vs $0.400 | – |
| Fastest first answer | Claude Opus 4.7 | 1.0 s (Claude Opus 4.7) vs 17.5 s (Grok 4) 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 58.1.
- Claude Opus 4.7 is stronger in agents & tools, coding, human preference, maths, multimodal, reasoning.
- Grok 4 is 1.7× cheaper ($6.00 vs $10.00 per 1M blended).
- Grok 4 streams 1.4× faster (69 vs 48 tokens per second).
At a glance
| Spec | Claude Opus 4.7 | Grok 4 |
|---|---|---|
| Developer | Anthropic | SpaceXAI |
| Released | 16 Apr 2026 | 9 Jul 2025 |
| Context window (tokens) | 1,000,000 | 256,000 |
| Compared at | Claude Opus 4.7 | Grok 4 |
| Open weights | No | No |
| Price in / out / cached input per 1M | $5.00 / $25.00 / $0.500 | $3.00 / $15.00 / – |
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 | Grok 4 | Leader |
|---|---|---|---|
| Reasoning | 63.6 | 60.0 | Claude Opus 4.7 |
| Coding | 60.9 | 55.9 | Claude Opus 4.7 |
| Agents & tools | 67.9 | 58.6 | Claude Opus 4.7 |
| Maths | 66.7 | 58.3 | Claude Opus 4.7 |
| Knowledge | – | 56.0 | not comparable |
| Instruction following | – | 62.5 | not comparable |
| Human preference | 67.3 | 58.5 | Claude Opus 4.7 |
| Multimodal | 67.1 | 52.0 | Claude Opus 4.7 |
| Long context | – | 68.3 | 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: Knowledge, Instruction following, Long context.
Where sources disagree
Across the 18 benchmarks both have, Claude Opus 4.7 leads 16, Grok 4 leads 2. Every overall index that rates both puts Claude Opus 4.7 ahead.
| Overall index | Claude Opus 4.7 | Grok 4 | Leader |
|---|---|---|---|
| BenchLeader Index | 63.1 | 58.1 | Claude Opus 4.7 |
| Arena Text | 1494 | 1411 | 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 | Grok 4 |
|---|---|---|
| Blended price per 1M tokens | $10.00 | $6.00 |
| Input / output per 1M | $5.00 / $25.00 | $3.00 / $15.00 |
| Cached input per 1M | $0.500 | – |
| Price per task: chat reply | $0.0095 | $0.0057 |
| Price per task: summarize a 30-page report | $0.075 | $0.045 |
| Price per task: code edit | $0.068 | $0.041 |
| Price per task: agentic coding session | $0.400 | $0.240 |
| Price per task: structured extraction | $0.015 | $0.0090 |
| Agentic coding session, input mostly cached | $0.198 | – |
| Output speed | 48 tok/s | 69 tok/s |
| Time to first answer | 1.0 s | 17.5 s |
| Time to a complete answer, last measured 13 Nov 2026 | – | 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. Grok 4 scores 58.1 and takes 18 s to its first answer token. 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
134 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 | Grok 4 |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 63.1 | 58.1 |
| Agents & tools score | BenchLeader | 67.9 | 58.6 |
| Coding score | BenchLeader | 60.9 | 55.9 |
| Human preference score | BenchLeader | 67.3 | 58.5 |
| Maths score | BenchLeader | 66.7 | 58.3 |
| Multimodal score | BenchLeader | 67.1 | 52.0 |
| Reasoning score | BenchLeader | 63.6 | 60.0 |
| Instruction following score | BenchLeader | – | 62.5 |
| Knowledge score | BenchLeader | – | 56.0 |
| Long context score | BenchLeader | – | 68.3 |
| Blended price $/M | BenchLeader | $10.00 | $6.00 |
| Output speed | BenchLeader | 48 tok/s | 69 tok/s |
| Time to first answer | BenchLeader | 1.0 s | 17.5 s |
| Context window | BenchLeader | 1,000,000 tokens | 256,000 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 86.4%† | 87.0% |
| Humanity's Last Exam | Scale AI / CAIS | 36.2% | – |
| SimpleBench | SimpleBench | 61.7% | 60.5% |
| Arena Hard Prompts | Arena (LMArena) | 1519 | 1421 |
| LiveBench 1.0 Reasoning | LiveBench | 87.2%† | – |
| GPQA Diamond (AA) | Artificial Analysis | 91.4%† | 87.7% |
| Humanity's Last Exam (AA) | Artificial Analysis | 42.3%† | 26.7% |
| GPQA Diamond (Vals) | Vals AI | 90.2%† | 88.1% |
| Kagi LLM Benchmark | Kagi LLM Benchmark | 73.3% | 73.6% |
| GPQA Diamond (HELM) | HELM Capabilities | – | 72.6% |
| ARC-AGI-1 | ARC Prize | – | 66.7%† |
| ARC-AGI-2 | ARC Prize | – | 16.0%† |
| ARC-AGI-3 | ARC Prize | 0.2%† | – |
| CritPt | Artificial Analysis | 12.0%† | 2.0% |
| Arena Expert | Arena (LMArena) | 1534 | 1429 |
| Chess Puzzles | Epoch AI Benchmarking Hub | 41.0%† | – |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 13.0% | – |
| LMCA | Epoch AI Benchmarking Hub | 52.2%† | – |
| DTBench | Epoch AI Benchmarking Hub | 94.7%† | – |
| ForecastBench | ForecastBench | 62.8% | 64.7% |
| Coding | |||
| SWE-bench Verified (Epoch) | Epoch AI Benchmarking Hub | 83.5%† | – |
| SciCode | SciCode | 54.5%† | – |
| WeirdML | WeirdML | 76.4% | 45.7% |
| FrontierCode | Cognition | 38.5% | – |
| GSO-Bench | GSO-Bench | 44.1% | – |
Show all 134 benchmarks
| Metric | Publisher | Claude Opus 4.7 | Grok 4 |
|---|---|---|---|
| Coding | |||
| Arena Coding | Arena (LMArena) | 1547 | 1435 |
| Arena WebDev | Arena (LMArena) | 1555 | – |
| LiveBench 1.0 Coding | LiveBench | 82.1%† | – |
| LiveCodeBench | Vals AI | 85.1%† | 83.3% |
| SWE-bench (Vals) | Vals AI | 82.0%† | 57.8% |
| Aider Polyglot | Aider polyglot leaderboard | – | 79.6%† |
| Code Migration | Vals AI | 43.9%† | – |
| ProgramBench | Vals AI | 0.0%† | – |
| Vibe Code Bench v1.1 | Vals AI | 71.0%† | – |
| SWE Atlas v1.0: Codebase QnA | Scale AI SEAL | 40.3% | – |
| SWE Atlas v1.0: Refactoring | Scale AI SEAL | 48.6% | – |
| SWE Atlas v1.0: Test Writing | Scale AI SEAL | 38.5% | – |
| MirrorCode | Epoch AI Benchmarking Hub | 31.1%† | – |
| ALE-Bench | Epoch AI Benchmarking Hub | 1323.0 | – |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | 83.2%† | – |
| Design Arena Website | Design Arena | – | 1006 |
| LiveCodeBench (AA) | Artificial Analysis | – | 81.9% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 80.2% | 27.2% |
| OSWorld-Verified 2.0 | OSWorld | 18.2%† | – |
| GDPval | OpenAI | – | 21.1%† |
| Cybench | Cybench | – | 43.0% |
| APEX-Agents | Mercor | 49.2%† | – |
| LiveBench Agentic Coding | LiveBench | 50.7%† | – |
| Terminal-Bench Hard | Artificial Analysis | 51.5%† | 37.9% |
| τ²-Bench Telecom (AA) | Artificial Analysis | 88.6%† | 74.8% |
| Terminal-Bench 2.1 (Vals) | Vals AI | 68.5%† | – |
| MCP Atlas | Scale AI SEAL | 79.1%† | – |
| HiL-Bench | Scale AI SEAL | 41.7% | – |
| BFCL Overall | Berkeley Function Calling Leaderboard | – | 63.0% |
| τ³-bench Banking | τ²-bench | 40.2%† | – |
| GDPval-AA v2.1 | Artificial Analysis | 1361† | – |
| τ³-Banking (AA) | Artificial Analysis | 34.6%† | – |
| ITBench SRE (AA) | Artificial Analysis | 46.7%† | – |
| Analyst Agent (AA) | Artificial Analysis | 43.8%† | – |
| Finance Agent v2 | Vals AI | 51.5%† | – |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 6.7%† | – |
| Legal Research Bench | Vals AI | 38.5%† | – |
| Tax Agent Bench | Vals AI | 23.7%† | – |
| Terminal-Bench 2.0 (Vals) | Vals AI | 68.5%† | 28.1% |
| Arena Search | Arena (LMArena) | 1210 | 1123 |
| BALROG | BALROG | – | 43.6% |
| DeepResearch Bench | DeepResearch Bench | – | 47.3% |
| PostTrainBench | Epoch AI Benchmarking Hub | 24.3%† | – |
| ExploitBench | Epoch AI Benchmarking Hub | 26.5% | – |
| METR Time Horizons | METR | – | 66.6% |
| GBAEval | Epoch AI Benchmarking Hub | 43.8% | – |
| Vending-Bench 2 | Andon Labs | 10936.8 | – |
| AA-Briefcase v1.1 | Artificial Analysis | 1252† | – |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 70.2%† | – |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 31.7%† | – |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 86.7%† | 84.0% |
| ProofBench | Vals AI | 54.0%† | – |
| LiveBench Mathematics | LiveBench | 92.8%† | – |
| AIME (Vals) | Vals AI | 96.3% | 90.6% |
| MGSM | Vals AI | – | 90.9% |
| AIME 2026 | MathArena | 95.8%† | – |
| HMMT February 2026 | MathArena | 93.9%† | – |
| IMO 2025 | MathArena | – | 21.4% |
| MathArena Apex | MathArena | 40.6%† | 2.1% |
| Omni-MATH (HELM) | HELM Capabilities | – | 60.3% |
| MATH 500 | Vals AI | – | 96.2% |
| Arena Maths | Arena (LMArena) | 1490 | 1425 |
| AIME 2025 (AA) | Artificial Analysis | – | 92.7% |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 51.7%† | – |
| LiveBench 1.0 Data Analysis | LiveBench | 78.3%† | – |
| MMLU-Pro | Vals AI | 89.9%† | 85.3% |
| LegalBench | Vals AI | 85.3%† | 83.2% |
| CorpFin | Vals AI | 66.1%† | 66.0% |
| TaxEval | Vals AI | 75.3%† | 65.1% |
| MedQA | Vals AI | – | 92.5% |
| MMLU-Pro (HELM) | HELM Capabilities | – | 85.1% |
| CaseLaw v2 | Vals AI | 68.4% | 65.8% |
| Excel Modeling Benchmark | Vals AI | 63.7%† | – |
| MedCode | Vals AI | 54.9%† | 38.1% |
| MedScribe | Vals AI | 83.0%† | 78.2% |
| EBR-bench | Epoch AI Benchmarking Hub | 19.1%† | – |
| GDP.pdf | Epoch AI Benchmarking Hub | 21.0%† | – |
| Global-MMLU-Lite (AA) | Artificial Analysis | – | 89.5% |
| Instruction following | |||
| LiveBench Language | LiveBench | 77.9%† | – |
| LiveBench Instruction Following | LiveBench | 66.7%† | – |
| IFBench | Artificial Analysis | 58.6%† | 53.7% |
| IFEval (HELM) | HELM Capabilities | – | 94.9% |
| WildBench (HELM) | HELM Capabilities | – | 79.7% |
| Arena Instruction Following | Arena (LMArena) | 1493 | 1396 |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1494 | 1411 |
| EQ-Bench 4 | EQ-Bench | 1293 | – |
| EQ-Bench Creative Writing v3 | EQ-Bench | 1915 | – |
| Arena Creative Writing | Arena (LMArena) | 1482 | 1394 |
| Arena Multi-turn | Arena (LMArena) | 1514 | 1413 |
| Arena Longer Queries | Arena (LMArena) | 1508 | 1416 |
| Lech Mazur Creative Writing | Lech Mazur | – | 76.9% |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1319 | 1210 |
| MMMU-Pro | Artificial Analysis | 78.8%† | 68.8% |
| MMMU-Pro (Vals) | Vals AI | 85.5%† | 76.3% |
| MortgageTax | Vals AI | 70.3%† | 44.5% |
| GeoBench | GeoBench | – | 45.0% |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 24.5% | – |
| Furniture Assembly | Epoch AI Benchmarking Hub | 33.3%† | – |
| Long context | |||
| Fiction.LiveBench 120k | Fiction.live | – | 96.9% |
| AA-LCR v1.1 | Artificial Analysis | 78.7%† | 68.0% |
| Arena Document | Arena (LMArena) | 1494 | – |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 157.4 | 146.9 |
| LiveBench 1.0 | LiveBench | 76.5%† | – |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 40.7† | 22.5 |
| AA Progress Index v1 | Artificial Analysis | 312.5† | 262.3 |
| HELM Capabilities mean | HELM Capabilities | – | 78.5% |
| Vals RSI Index | Vals AI | 19.1%† | – |
| Vals Multimodal Index | Vals AI | 67.4%† | – |
| Metadata | |||
| AA Openness Index | Artificial Analysis | – | 5.6 |
Data as of 2026-11-16. Best configuration of each model; every score links to its source on the model pages.
What changed
- Grok 4: 1 new result (AIME 2025 (AA)).
- Grok 4: 1 new result (Arena Search).
- Claude Opus 4.7: 1 score revision.
- Grok 4: 1 score revision.
- Grok 4: BenchLeader Index 58.8 → 58.3.
- Grok 4: 1 new result (AA Progress Index v1).
- Claude Opus 4.7: 1 score revision.
- Grok 4: 1 new result (Global-MMLU-Lite (AA)).
- Claude Opus 4.7: 1 new result (Harvey's Legal Agent Benchmark v1).
- 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
- Grok 4 vs Gemini 4 Carbon
- Grok 4 vs GPT-6 Astra
- 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 Grok 4: questions
- Is Claude Opus 4.7 better than Grok 4?
- Claude Opus 4.7 leads on quality: 63.1 vs 58.1. 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 Grok 4: which is cheaper, and will that change?
- Grok 4 is cheaper today: $6.00 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 Grok 4 better for coding?
- Claude Opus 4.7 scores higher in coding (61 vs 56 on the category index, where 50 is average). On WeirdML, the top shared coding benchmark, it is 76.4% for Claude Opus 4.7 against 45.7% for Grok 4.
- Which is faster, Claude Opus 4.7 or Grok 4?
- Grok 4 streams ~1.5× faster (69 vs 48 output tokens per second, Artificial Analysis 7-day median).
- Is Claude Opus 4.7 or Grok 4 better for agents and tool use?
- Claude Opus 4.7 scores higher in agentic tasks (68 vs 59 on the category index, where 50 is average). On Terminal-Bench, the top shared agents & tools benchmark, it is 80.2% for Claude Opus 4.7 against 27.2% for Grok 4.
More questions (4)
- Which is better for long documents?
- Claude Opus 4.7 accepts more context: 1,000,000 against 256,000 tokens for Grok 4.
- Which has lower latency?
- Claude Opus 4.7 answers first: 1.0 s against 17.5 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 Grok 4?
- Across the 18 benchmarks both have, Claude Opus 4.7 leads 16, Grok 4 leads 2. Every overall index that rates both puts Claude Opus 4.7 ahead.
- When was Claude Opus 4.7 released?
- Claude Opus 4.7 was released on 16 Apr 2026.