AnthropicClaude Opus 4.8
Claude Opus 4.8 is an Anthropic proprietary reasoning model, released 28 May 2026. Its best configuration (max reasoning effort) ranks #64 of 775 on the BenchLeader Index at 63.4 ±4.8, in the upper half. The ± is the point: 160 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in composite (76) and lowest in agents & tools (57). At $10.00 per million tokens blended it is among the most expensive ranked models. Output speed of 63 tokens per second puts it slower than most, with a first answer in 44.6 s. It has been measured at 8 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 1 Jun 2026.
- Blended price
- $10.00/M
- $5.00 in · $25.00 out
- Output speed
- 63 tok/s
- measured by Artificial Analysis
- First answer
- 45 s
- first token 0.97 s
- Context
- 1M
- Full answer
- 53 s
- median, reasoning included
- Cost per run
- $4.08
- one full Intelligence Index run
- Released
- 28 May 2026
- Overall index63
- Reasoning69
- Coding59
- Agents & tools57
- Maths68
- Knowledge67
- Instruction following62
- Long context64
- Composite76
Versions
Anthropic has shipped 9 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| Claude Opus 5.5max | 22 Sept 2026 | 72.7 | #1 |
| Claude Opus 5max | 24 Jul 2026 | 68.9 | #15 |
| Claude Opus 4.8maxthis page | 28 May 2026 | 63.4 | #64 |
| Claude Opus 4.7 | 16 Apr 2026 | 63.4 | #63 |
| Claude Opus 4.6high | 5 Feb 2026 | 62.9 | #76 |
| Claude Opus 4.5thinking | 24 Nov 2025 | 60.0 | #119 |
| Claude Opus 4.1thinking | 5 Aug 2025 | 56.5 | #199 |
| Claude Opus 4thinking | 22 May 2025 | 55.1 | #230 |
| Claude 3 Opus | 4 Mar 2024 | 39.2 | #681 |
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| no reasoning | – | – | 66 tok/s | 0.97 s | $0.0095 | Coding 64 · Maths 60 · Reasoning 62 |
| low | – | – | 66 tok/s | 0.97 s | $0.0095 | Maths 66 · Reasoning 62 |
| medium | – | – | 66 tok/s | 0.97 s | $0.0095 | Coding 68 · Reasoning 64 |
| high | 60.9 | #104 | 66 tok/s | 0.97 s | $0.0095 | Coding 62 · Human preference 68 · Maths 68 · Multimodal 64 · Reasoning 61 |
| xhigh | – | – | 66 tok/s | 0.97 s | $0.0095 | Coding 72 · Reasoning 30 |
| maxbest | 63.4 | #64 | 63 tok/s | 45 s | $0.0095 | Agents & tools 57 · Coding 59 · Composite 76 · Instruction following 62 · Knowledge 67 · Long context 64 · Maths 68 · Reasoning 69 |
| thinking | – | – | 66 tok/s | 0.97 s | $0.0095 | Reasoning 80 |
| not stated | 60.1 | #117 | 66 tok/s | 0.97 s | $0.0095 | Agents & tools 52 · Coding 62 · Composite 60 · Human preference 65 · Knowledge 62 · Maths 66 · Multimodal 63 · Reasoning 64 |
How its index has moved
8 Sept 2026 to 21 Oct 2026- max
- high
- not stated
The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | no reasoning | low | medium | high | xhigh | max | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|---|
| GPQA Diamond | 85.3%#87 | 88.4%#59 | – | – | – | 91.0%#32 | – | – | Epoch AI Benchmarking Hub | |
| SimpleBencheffort not stated | – | – | – | – | – | – | – | 64.8%#17 | SimpleBench | |
| Arena Hard Prompts | – | – | – | 1513#15 | – | – | – | 1505#23 | Arena (LMArena) | |
| LiveBench 2.0 Reasoningnot in indexeffort not stated | – | – | – | – | – | – | – | 89.2%#18 | LiveBench | |
| LiveBench 1.0 Reasoningnot in index | – | – | – | – | – | 89.2%#19 | – | – | LiveBench | |
| GPQA Diamond (AA)not in index | – | – | – | – | – | 92.0%#41 | – | – | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | – | – | – | – | – | 48.7%#32 | – | – | Artificial Analysis | |
| GPQA Diamond (Vals)not in indexeffort not stated | – | – | – | – | – | – | – | 92.4%#16 | Vals AI | |
| EnigmaEval | – | – | – | – | 23.5%#5 | – | – | – | Scale AI SEAL | |
| Kagi LLM Benchmark | – | – | – | – | – | – | 88.8%#2 | – | Kagi LLM Benchmark | |
| ARC-AGI-1 | – | 88.0%#83 | 91.5%#60 | 92.0%#56 | – | 92.5%#51 | – | – | ARC Prize | |
| ARC-AGI-2 | – | 62.2%#73 | 71.7%#54 | 72.1%#52 | – | – | – | – | ARC Prize | |
| ARC-AGI-3 | – | – | – | 1.5%#35 | – | – | – | – | ARC Prize | |
| CritPt | – | – | – | – | – | 20.9%#56 | – | – | Artificial Analysis | |
| Chess Puzzlesnot in index | 19.0%#94 | 29.0%#68 | – | – | – | 52.0%#17 | – | – | Epoch AI Benchmarking Hub | |
| Mystery Game Puzzlesnot in index | – | – | – | – | 31.0%#32 | 36.0%#21 | – | – | Epoch AI Benchmarking Hub | |
| LMCAnot in index | – | – | – | – | – | 57.5%#39 | – | – | Epoch AI Benchmarking Hub | |
| DTBenchnot in index | – | – | – | – | – | 94.9%#46 | – | – | Epoch AI Benchmarking Hub | |
| ForecastBenchnot in index | – | – | – | – | – | 66.3%#4 | – | 62.6%#53 | ForecastBench | |
| BTF-3not in index | – | – | – | 14.0%#6 | 13.2%#11 | – | – | – | Epoch AI Benchmarking Hub |
Coding
| Benchmark | no reasoning | low | medium | high | xhigh | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| SciCode | – | – | – | – | – | 53.5%#75 | – | SciCode | |
| WeirdML | 70.5%#32 | – | 76.0%#27 | – | 82.9%#17 | – | – | WeirdML | |
| FrontierCodeeffort not stated | – | – | – | – | – | – | 46.5%#14 | Cognition | |
| GSO-Bencheffort not stated | – | – | – | – | – | – | 47.1%#5 | GSO-Bench | |
| Arena Coding | – | – | – | 1534#17 | – | – | 1529#23 | Arena (LMArena) | |
| Arena WebDev | – | – | – | 1556#38 | – | – | 1536#47 | Arena (LMArena) | |
| LiveBench 2.0 Codingnot in indexeffort not stated | – | – | – | – | – | – | 77.3%#33 | LiveBench | |
| LiveBench 1.0 Codingnot in index | – | – | – | – | – | 81.8%#12 | – | LiveBench | |
| SciCode (AA)not in index | – | – | – | – | – | 54.4%#62 | – | Artificial Analysis | |
| LiveCodeBencheffort not stated | – | – | – | – | – | – | 87.8%#11 | Vals AI | |
| SWE-bench (Vals)not in indexeffort not stated | – | – | – | – | – | – | 88.6%#12 | Vals AI | |
| Code Migrationnot in indexeffort not stated | – | – | – | – | – | – | 47.3%#15 | Vals AI | |
| ProgramBenchnot in indexeffort not stated | – | – | – | – | – | – | 1.0%#15 | Vals AI | |
| Vibe Code Bench v1.1not in indexeffort not stated | – | – | – | – | – | – | 82.7%#21 | Vals AI | |
| SWE Atlas: Codebase QnAnot in index | – | – | – | – | 57.3%#4 | – | – | Scale AI SEAL | |
| SWE Atlas: Refactoringnot in indexeffort not stated | – | – | – | – | – | – | 46.7%#5 | Scale AI SEAL | |
| SWE Atlas: Test Writingnot in index | – | – | – | – | 49.6%#6 | – | – | Scale AI SEAL | |
| Surface Evolver Benchnot in index | 68.1%#11 | – | – | 87.5%#6 | – | – | – | Epoch AI Benchmarking Hub | |
| DeepSWE v1.1not in index | – | 40.8%#56 | 48.7%#49 | 51.8%#46 | 54.4%#39 | 59.0%#34 | – | Epoch AI Benchmarking Hub | |
| ALE-Benchnot in index | 1411.8#26 | – | – | 1563.8#21 | – | – | – | Epoch AI Benchmarking Hub | |
| Terminal-Bench 4.0 (AA)not in index | – | – | – | – | – | 21.7%#58 | – | Artificial Analysis | |
| Terminal-Bench 2.1 (AA)not in index | – | – | – | – | – | 84.6%#28 | – | Artificial Analysis |
Agents & tools
| Benchmark | low | medium | high | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|
| Terminal-Bench | – | – | – | 23.6%#68 | – | Terminal-Bench | |
| OSWorld-Verified 2.0 | – | – | – | 20.6%#7 | – | OSWorld | |
| Remote Labor Indexeffort not stated | – | – | – | – | 8.3%#5 | Scale AI / CAIS | |
| APEX-Agents | – | – | – | 48.9%#29 | – | Mercor | |
| LiveBench Long Agenticnot in indexeffort not stated | – | – | – | – | 48.3%#16 | LiveBench | |
| LiveBench Browsernot in indexeffort not stated | – | – | – | – | 96.0%#2 | LiveBench | |
| LiveBench Agentic Codingnot in index | – | – | – | 50.5%#43 | – | LiveBench | |
| Terminal-Bench Hard | – | – | – | 58.3%#10 | – | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | – | – | – | 94.4%#30 | – | Artificial Analysis | |
| Terminal-Bench 2.1 (Vals)effort not stated | – | – | – | – | 71.9%#25 | Vals AI | |
| MCP Atlas | – | – | – | 82.2%#9 | – | Scale AI SEAL | |
| HiL-Bencheffort not stated | – | – | – | – | 35.3%#21 | Scale AI SEAL | |
| τ³-bench Bankingnot in index | – | – | – | 39.7%#8 | – | τ²-bench | |
| GDPval-AA v2.1not in index | – | – | – | 47.7%#60 | – | Artificial Analysis | |
| τ³-Banking (AA) | – | – | – | 34.2%#54 | – | Artificial Analysis | |
| Analyst Agent (AA)not in index | – | – | – | 45.0%#17 | – | Artificial Analysis | |
| Finance Agent v2not in indexeffort not stated | – | – | – | – | 53.9%#28 | Vals AI | |
| Legal Research Benchnot in indexeffort not stated | – | – | – | – | 43.8%#20 | Vals AI | |
| SkillsBenchnot in indexeffort not stated | – | – | – | – | 59.2%#9 | Vals AI | |
| Tax Agent Benchnot in indexeffort not stated | – | – | – | – | 30.0%#28 | Vals AI | |
| Terminal-Bench 2.0 (Vals)not in indexeffort not stated | – | – | – | – | 70.0%#2 | Vals AI | |
| Terminal-Bench 4.0 (Vals)not in indexeffort not stated | – | – | – | – | 23.2%#25 | Vals AI | |
| Terminal-Bench Sciencenot in indexeffort not stated | – | – | – | – | 4.3%#25 | Vals AI | |
| Time Horizon Index: KSPnot in indexeffort not stated | – | – | – | – | 11.8%#8 | Vals AI | |
| Arena Searchnot in indexeffort not stated | – | – | – | – | 1195#22 | Arena (LMArena) | |
| DeepResearch Benchnot in index | 49.3%#14 | 47.4%#24 | 50.2%#10 | – | – | DeepResearch Bench | |
| PostTrainBenchnot in index | – | – | 33.8%#4 | 32.9%#5 | – | Epoch AI Benchmarking Hub | |
| GBAEvalnot in indexeffort not stated | – | – | – | – | 70.9%#3 | Epoch AI Benchmarking Hub | |
| Vending-Bench 2not in index | – | – | – | 2992.3#49 | 5787.4#23 | Andon Labs | |
| AutomationBenchnot in index | – | – | – | 45.6%#89 | – | Artificial Analysis | |
| GDP.pdf (AA)not in index | – | – | – | 22.8%#44 | – | Artificial Analysis | |
| EnterpriseOps-Gymnot in index | – | – | – | 44.0%#18 | – | Artificial Analysis | |
| FACTS Searchnot in indexeffort not stated | – | – | – | – | 79.6%#13 |
Maths
| Benchmark | no reasoning | low | high | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|
| FrontierMath Tiers 1–3 | – | – | – | 80.0%#17 | – | Epoch AI Benchmarking Hub | |
| FrontierMath Tier 4 | – | – | – | 56.1%#24 | – | Epoch AI Benchmarking Hub | |
| OTIS Mock AIME | 84.4%#97 | 97.8%#30 | – | 98.3%#23 | – | Epoch AI Benchmarking Hub | |
| ProofBench | – | – | – | 69.0%#17 | – | Vals AI | |
| LiveBench Mathematicsnot in index | – | – | – | 94.3%#19 | – | LiveBench | |
| AIME 2026possible contamination | – | – | – | 100.0%#1 | – | MathArena | |
| HMMT February 2026possible contamination | – | – | – | 95.5%#6 | – | MathArena | |
| MathArena Apexpossible contamination | – | – | – | 81.3%#1 | – | MathArena | |
| Arena Maths | – | – | 1494#25 | – | 1476#43 | Arena (LMArena) |
Knowledge
Instruction following
| Benchmark | high | max | not stated | Source | Trend |
|---|---|---|---|---|---|
| LiveBench Languagenot in index | – | 79.7%#35 | – | LiveBench | |
| LiveBench Instruction Followingnot in index | – | 72.0%#20 | – | LiveBench | |
| IFBench | – | 62.2%#117 | – | Artificial Analysis | |
| Arena Instruction Followingnot in index | 1491#12 | – | 1478#22 | Arena (LMArena) |
Human preference
| Benchmark | high | not stated | Source | Trend |
|---|---|---|---|---|
| Arena Text | 1482#25 | 1474#39 | Arena (LMArena) | |
| EQ-Bench 4effort not stated | – | 1281#6 | EQ-Bench | |
| Arena Creative Writingnot in index | 1469#18 | 1463#26 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1497#12 | 1495#17 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1503#10 | 1499#16 | Arena (LMArena) |
Multimodal
| Benchmark | high | not stated | Source | Trend |
|---|---|---|---|---|
| Arena Vision | 1295#23 | 1289#28 | Arena (LMArena) | |
| LiveBench Designnot in indexeffort not stated | – | 78.8%#16 | LiveBench | |
| MMMU-Pro (Vals)not in indexeffort not stated | – | 86.6%#18 | Vals AI | |
| MortgageTaxnot in indexeffort not stated | – | 69.9%#5 | Vals AI | |
| SAGEnot in indexeffort not stated | – | 54.8%#3 | Vals AI | |
| Blueprint-Bench 2not in indexeffort not stated | – | 14.5%#26 | Epoch AI Benchmarking Hub |
Long context
| Benchmark | high | max | not stated | Source | Trend |
|---|---|---|---|---|---|
| AA-LCR | – | 77.7%#127 | – | Artificial Analysis | |
| Arena Documentnot in index | 1486#9 | – | 1475#16 | Arena (LMArena) | |
| MLCRnot in index | – | 45.6%#14 | – | Artificial Analysis |
Composite
| Benchmark | max | not stated | Source | Trend |
|---|---|---|---|---|
| Epoch Capabilities Indexnot in indexeffort not stated | – | 159.0#17 | Epoch AI Benchmarking Hub | |
| LiveBench 2.0effort not stated | – | 76.8%#13 | LiveBench | |
| LiveBench 1.0not in index | 76.2%#30 | – | LiveBench | |
| AA Intelligence Index v4.3.2 | 41.8#56 | – | Artificial Analysis | |
| Vals Multimodal Indexnot in indexeffort not stated | – | 70.9%#5 | Vals AI |
Safety & honesty
| Benchmark | max | Source | Trend |
|---|---|---|---|
| FORTRESSnot in index | 18.2%#42 | Scale AI SEAL |
Where it wins
Benchmarks where this configuration ranks in the top five of every configuration measured.
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Anthropic | 159 tok/s | 0.58 s | $10.00 | $50.00 | 1M | – |
| Google Vertex | 78 tok/s | 2.23 s | $5.00 | $25.00 | 1M | – |
| Amazon Bedrock | 75 tok/s | 2.51 s | $5.00 | $25.00 | 1M | – |
| Anthropic | 73 tok/s | 2.90 s | $5.00 | $25.00 | 1M | – |
| Amazon Bedrock | 66 tok/s | 3.17 s | $5.50 | $27.50 | 1M | – |
| Claude Platform on AWS | 65 tok/s | 1.02 s | $5.00 | $25.00 | 1M | – |
| Azure | 41 tok/s | 1.12 s | $5.00 | $25.00 | 1M | – |
Price history
Listed price per 1M tokens over time, as recorded by OpenRouter for the provider with the longest history.
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.500 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0095 | $0.0081 | 49.3 s |
| Summarize a 30-page report | 12,000 / 600 | $0.075 | $0.035 | 54.1 s |
| Code edit | 6,000 / 1,500 | $0.068 | $0.047 | 1.1 min |
| Agentic coding session | 60,000 / 4,000 | $0.400 | $0.198 | 1.8 min |
| Structured extraction | 2,000 / 200 | $0.015 | $0.0083 | 47.7 s |
More from Anthropic
Compare Claude Opus 4.8 with…
- Claude Opus 4.8 vs Claude Opus 5.5
- Claude Opus 4.8 vs Gemini 4 Argon
- Claude Opus 4.8 vs Muse Wonder Full
- Claude Opus 4.8 vs GPT-6 Astra
- Claude Opus 4.8 vs Claude Fable 5.1
- Claude Opus 4.8 vs GPT-6.1 Sol
- Claude Opus 4.8 vs Claude Sonnet 5.5
- Claude Opus 4.8 vs Claude Fable 5
- Claude Opus 4.8 vs Claude Opus 5
- Claude Opus 4.8 vs GPT-5.6 Sol
- Claude Opus 4.8 vs Muse Wonder
- Claude Opus 4.8 vs Muse Spark 1.3
- Claude Opus 4.8 vs GPT-6 Sol
- Claude Opus 4.8 vs GPT-5.5
- Claude Opus 4.8 vs GPT-5.5 Pro
- Claude Opus 4.8 vs Kimi K3
- Claude Opus 4.8 vs GPT-5.4
- Claude Opus 4.8 vs Gemini 3.8 Flash
- Claude Opus 4.8 vs Muse Spark
- Claude Opus 4.8 vs GPT-5.6 Terra
- Claude Opus 4.8 vs MiMo-V2.6-Pro
- Claude Opus 4.8 vs Grok 4.6
- Claude Opus 4.8 vs Grok 4.7
- Claude Opus 4.8 vs GPT-5.3-Codex
- Claude Opus 4.8 vs GLM 5.3
- Claude Opus 4.8 vs Gemini 3.7 Flash
- Claude Opus 4.8 vs GPT-5.4 Pro
- Claude Opus 4.8 vs Claude Opus 4.7
- Claude Opus 4.8 vs Qwen3.8 Max
- Claude Opus 4.8 vs Gemini 3.5 Flash
- Claude Opus 4.8 vs Muse Spark 1.2
- Claude Opus 4.8 vs Qwen3.8 2.4T A95B
- Claude Opus 4.8 vs DeepSeek V4.1 Flash 1017
- Claude Opus 4.8 vs Muse Spark 1.1
- Claude Opus 4.8 vs Claude Opus 4.6
- Claude Opus 4.8 vs GLM 5.3 Flash
- Claude Opus 4.8 vs DeepSeek V4 Pro
- Claude Opus 4.8 vs GLM 5.2
- Claude Opus 4.8 vs Qwen3.6 Max
See also
Data as of 21 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “Claude Opus 4.8: benchmarks, pricing, speed and rank”, https://www.benchleader.com/models/claude-opus-4-8, data as of 21 Oct 2026.