Models / Claude Opus 4.5

Anthropic

Claude Opus 4.5

Claude Opus 4.5 is best for writing, and agents and tool use; pick it when quality matters more than cost. The cheapest model of similar intelligence is MiMo-V2.6-Flash, at $0.18 per million tokens against $10.00.

Superseded. Claude Opus 5.5 is Anthropic's newer model in this line, at $8.00 blended: see Claude Opus 5.5 or compare the two.

Results 16 Nov 2026 · 71 results · 10 sources

Test Claude Opus 4.5 on your prompt → It runs beside the models you could use instead, with every answer checked and costed.

Price per million tokens
$5.00 in · $25.00 out
Typical host on OpenRouter, 10 Nov 2026 · compare prices
Speed
Not measured
Intelligence Index
29.1
Artificial Analysis, at thinking reasoning; 23.7–29.1 across 2 settings
Spring Prompt overall
Not ranked yet
needs our benchmarks and two groups of results
Developer
Anthropic, based in the US
Weights not published
Released
24 Nov 2025
200,000 tokens of context

Where it stands among the models you could choose

Available models grouped by price, intelligence, speed and developer location. Retired models are excluded.

AmongIntelligence IndexSpring Prompt overallPrice, blendedSpeed
Models available today 57th of 185Claude Opus 5.5 – $10.00 blended, 94th percentilecheapest: Mistral Nemo –
Models priced $10 or more per million tokens 9th of 11Claude Fable 5.1 – sets this group –
Models with similar intelligence sets this group – $10.00 blended, 96th percentilecheapest: Nemotron 3.5 Super VL –
Models from developers based in the US 37th of 92Claude Opus 5.5 – $10.00 blended, 88th percentilecheapest: Gemma 3 4B –

The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.

How good is it, and for what?

Each use case ranks the available models its benchmarks measured, then those priced $10 or more per million tokens. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.

Product listingsTurning a sparse product feed and photos into listings that can go live Not measured
  • CatalogBench: has not measured this model.
Decks from an analysisTurning a finished analysis into a deck you could present as it is Not measured
  • DeckBench: has not measured this model.
User surveysPlanning a user survey and reading its results without being misled Not measured
  • SurveyBench: has not measured this model.
Marketing planningPlanning a year of ad spend without overspending Not measured
  • ROASBench: has not measured this model.
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 36th of 159 available · 9th of 9 at its price
BenchmarkClaude Opus 4.5Rank among availableBest available
tau2-benchtau2 v1.0.1, retail: task success (pass^1) · reported 79.6% 3rd of 6 Qwen3.5-397B-A17B 84.4%
tau2-benchtau2 v1.0.1, airline: task success (pass^1) · reported 84.0% 1st of 6 This model
tau2-benchtau2 v1.0.1, telecom: task success (pass^1) · reported 92.3% 2nd of 6 Qwen3.5-397B-A17B 97.8%
Berkeley Function Calling Leaderboard (BFCL) V4BFCL: overall accuracy · reported 77.5% 1st of 36 This model
OpenHands IndexOpenHands Index: average · reported 60.6% 8th of 31 Claude Fable 5 81.0%
Vending-Bench 2Vending-Bench 2: money after a year · reported $4,967 27th of 53 GPT-6 Astra $15,515
  • Microsoft STATE-Bench: has not measured this model.
  • Artificial Analysis: has not measured this model.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects 28th of 51 available · 9th of 9 at its price
BenchmarkClaude Opus 4.5Rank among availableBest available
Remote Labor IndexRemote Labor Index: projects done to client standard · reported 3.8% 8th of 12 GPT-6 Astra 20.8%
  • APEX-Agents: has not measured this model.
  • GDP.pdf: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge 72nd of 187 available · 9th of 11 at its price
BenchmarkClaude Opus 4.5Rank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 29.1 57th of 185 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 30.1% 74th of 186 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 86.6% 62nd of 168 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 18th of 169 available · 5th of 10 at its price
BenchmarkClaude Opus 4.5Rank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,474 28th of 153 Gemini 4 Argon 1,525
UGI LeaderboardUGI: writing score · reported 70.4 7th of 124 Gemini 3.8 Flash 78.6
Sticking to the factsSummarising without inventing things, and factual answers 67th of 139 available · 10th of 12 at its price
BenchmarkClaude Opus 4.5Rank among availableBest available
Vectara Hallucination LeaderboardVectara: hallucination rate · reported 10.9% 49th of 71 GPT-5.4 nano 3.1%
Arena (formerly LMArena)Arena Text factuality: overall · reported 1,459 37th of 109 Gemini 4 Argon 1,511
SimpleQA Verified (Epoch AI)SimpleQA Verified: correct answers · reported 45.7% 33rd of 68 GPT-6 Astra 75.6%
Speed under pressureGood decisions against a real clock (fast chess) Not measured
  • BulletBench: has not measured this model.

Nearest alternatives

Against Claude Opus 4.5 at $10.00 blended, Intelligence Index 29.1. "Similar intelligence" means within 4 points or better.

Where to run it

6 hosts on OpenRouter at the standard service tier, with a median of $10.00 blended per million tokens. Cheapest first.

HostQuantisationInputOutputBlended
Amazon Bedrocknot stated$5.00$25.00$10.00
Anthropicnot stated$5.00$25.00$10.00
Azurenot stated$5.00$25.00$10.00
Claude Platform on AWSnot stated$5.00$25.00$10.00
Googlenot stated$5.00$25.00$10.00
Amazon Bedrocknot stated$5.50$27.50$11.00

Prices per million tokens from OpenRouter's public endpoint list, checked 10 Nov 2026.

Against the alternatives

The models you would most likely weigh it against: the leaders of the groups above.

Scroll sideways to see every alternative.

Claude Opus 4.5Claude Fable 5.1
most intelligent at $10 or more per million tokens
Nemotron 3.5 Super VL
cheapest with similar intelligence
Claude Opus 5.5
most intelligent available today
Price per million tokens $10.00 $20.00$0.24$8.00
Tokens a second – 7241196
Intelligence Index 29.1 53.426.557.6
Spring Prompt overall – 64–82
Rank among available models, by use case
Product listings – 24th–13th
Decks from an analysis – 11th–7th
User surveys – ––4th
Marketing planning – 10th–9th
Agents and tool use 36th of 159 19th132nd3rd
Professional work 28th of 51 4th–3rd
Reasoning and knowledge 72nd of 187 5th74th1st
Writing 18th of 169 19th–5th
Sticking to the facts 67th of 139 5th–3rd
Speed under pressure – –––

Shaded figures are better than Claude Opus 4.5's. A dash means no result.

Research and examples

Where it has been measured

9th most intelligent of the 11 models priced $10 or more per million tokens. Among models available today, in the top quarter (by percentile) for writing, and agents and tool use.

Artificial Analysis lists it but has not published a speed for it yet

Every result

71 published results from 10 sources, each in the source's own units, with the configuration that produced it.

Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 16 results
MeasureValueRankConfigurationDated
Arena Document: overallArena rating, higher is better 1,465
1,454–1,475
21 of 44 Claude Opus 4.5 13 Sep 2026
Arena Search: overallArena rating, higher is better 1,187
1,181–1,193
20 of 34 Claude Opus 4.5 (arena search) 24 Aug 2026
Arena Text factuality: overallArena rating, higher is better 1,459
1,456–1,462
48 of 182 Claude Opus 4.5 2 Oct 2026
Arena Text factuality: overallArena rating, higher is better 1,459
1,456–1,463
49 of 182 Claude Opus 4.5 (high-32k reasoning) 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,472
1,466–1,478
37 of 406 Claude Opus 4.5 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,467
1,460–1,475
43 of 406 Claude Opus 4.5 (high-32k reasoning) 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,462
1,456–1,468
25 of 411 Claude Opus 4.5 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,470
1,461–1,478
16 of 411 Claude Opus 4.5 (high-32k reasoning) 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,503
1,494–1,511
39 of 363 Claude Opus 4.5 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,503
1,491–1,515
38 of 363 Claude Opus 4.5 (high-32k reasoning) 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,475
1,470–1,480
27 of 413 Claude Opus 4.5 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,483
1,476–1,490
19 of 413 Claude Opus 4.5 (high-32k reasoning) 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,470
1,467–1,473
47 of 413 Claude Opus 4.5 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,474
1,470–1,477
39 of 413 Claude Opus 4.5 (high-32k reasoning) 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,467
1,461–1,472
27 of 412 Claude Opus 4.5 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,466
1,459–1,473
30 of 412 Claude Opus 4.5 (high-32k reasoning) 2 Oct 2026

Not shown: Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.

Artificial Analysis · reported by Artificial Analysis · 14 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 23.7 169 of 413 Claude Opus 4.5 (no reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 29.1 124 of 413 Claude Opus 4.5 (thinking reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 81.0% 168 of 360 Claude Opus 4.5 (no reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 86.6% 109 of 360 Claude Opus 4.5 (thinking reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 13.2% 249 of 419 Claude Opus 4.5 (no reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 30.1% 146 of 419 Claude Opus 4.5 (thinking reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 43.0% 184 of 288 Claude Opus 4.5 (no reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 58.0% 111 of 288 Claude Opus 4.5 (thinking reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 70.7% 193 of 407 Claude Opus 4.5 (no reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 77.3% 128 of 407 Claude Opus 4.5 (thinking reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 40.9% 51 of 282 Claude Opus 4.5 (no reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 47.0% 26 of 282 Claude Opus 4.5 (thinking reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 86.3% 72 of 286 Claude Opus 4.5 (no reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 89.5% 58 of 286 Claude Opus 4.5 (thinking reasoning) 20 Oct 2026

Not shown: Not business work, and a blend: read the parts for any one task.

Berkeley Function Calling Leaderboard (BFCL) V4 · reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 16 results
MeasureValueRankConfigurationDated
BFCL: irrelevance detection% correct, higher is better 84.7% 35 of 109 Claude Opus 4.5 (bfcl fc) 16 Dec 2025
BFCL: irrelevance detection% correct, higher is better 90.8% 14 of 109 Claude Opus 4.5 (bfcl prompt) 16 Dec 2025
BFCL: memory% correct, higher is better 73.8% 1 of 109 Claude Opus 4.5 (bfcl fc) 16 Dec 2025
BFCL: memory% correct, higher is better 1.9% 102 of 109 Claude Opus 4.5 (bfcl prompt) 16 Dec 2025
BFCL: multi-turn tasks% correct, higher is better 68.4% 4 of 109 Claude Opus 4.5 (bfcl fc) 16 Dec 2025
BFCL: multi-turn tasks% correct, higher is better 16.1% 56 of 109 Claude Opus 4.5 (bfcl prompt) 16 Dec 2025
BFCL: overall accuracy% correct, higher is better 77.5% 1 of 109 Claude Opus 4.5 (bfcl fc) 16 Dec 2025
BFCL: overall accuracy% correct, higher is better 33.5% 57 of 109 Claude Opus 4.5 (bfcl prompt) 16 Dec 2025
BFCL: relevance detection% correct, higher is better 62.5% 91 of 109 Claude Opus 4.5 (bfcl fc) 16 Dec 2025
BFCL: relevance detection% correct, higher is better 68.8% 80 of 109 Claude Opus 4.5 (bfcl prompt) 16 Dec 2025
BFCL: single-turn calls (curated)% correct, higher is better 88.6% 15 of 109 Claude Opus 4.5 (bfcl fc) 16 Dec 2025
BFCL: single-turn calls (curated)% correct, higher is better 89.7% 5 of 109 Claude Opus 4.5 (bfcl prompt) 16 Dec 2025
BFCL: single-turn calls (user-contributed)% correct, higher is better 79.8% 14 of 109 Claude Opus 4.5 (bfcl fc) 16 Dec 2025
BFCL: single-turn calls (user-contributed)% correct, higher is better 76.0% 35 of 109 Claude Opus 4.5 (bfcl prompt) 16 Dec 2025
BFCL: web search% correct, higher is better 84.5% 1 of 109 Claude Opus 4.5 (bfcl fc) 16 Dec 2025
BFCL: web search% correct, higher is better 13.0% 52 of 109 Claude Opus 4.5 (bfcl prompt) 16 Dec 2025

Not shown: Not general refusal or safety behaviour.

OpenHands Index · reported by OpenHands Index · 6 results
MeasureValueRankConfigurationDated
OpenHands Index: average% resolved, higher is better 60.6% 8 of 34 Claude Opus 4.5 (openhands) 30 Jun 2026
OpenHands Index: front end (SWE-Bench Multimodal)% resolved, higher is better 41.2% 6 of 34 Claude Opus 4.5 (openhands) 30 Jun 2026
OpenHands Index: greenfield (Commit0)% resolved, higher is better 37.5% 10 of 34 Claude Opus 4.5 (openhands) 30 Jun 2026
OpenHands Index: information gathering (GAIA)% resolved, higher is better 69.1% 13 of 34 Claude Opus 4.5 (openhands) 30 Jun 2026
OpenHands Index: issue resolution (SWE-Bench)% resolved, higher is better 76.6% 8 of 34 Claude Opus 4.5 (openhands) 30 Jun 2026
OpenHands Index: testing (SWT-Bench)% resolved, higher is better 78.5% 7 of 34 Claude Opus 4.5 (openhands) 30 Jun 2026

Not shown: Not the model alone: the score is for the model inside the OpenHands agent, at the SDK version shown.

Remote Labor Index · reported by Remote Labor Index · 1 result
MeasureValueRankConfigurationDated
Remote Labor Index: projects done to client standard% of projects, higher is better 3.8% 8 of 14 Claude Opus 4.5 1 Oct 2026

Not shown: Not speed or cost; a small set of projects, so a few points either way are noise.

SimpleQA Verified (Epoch AI) · reported by SimpleQA Verified (Epoch AI) · 1 result
MeasureValueRankConfigurationDated
SimpleQA Verified: correct answers% of questions, higher is better 45.7% 37 of 85 Claude Opus 4.5 (thinking-32k reasoning) 27 Aug 2026

Not shown: Not answers grounded in your documents; tests what the model remembers.

tau2-bench · reported by tau2-bench · 8 results
MeasureValueRankConfigurationDated
tau2 v1.0.1, airline: consistency (pass^4)% of tasks, higher is better 70.0% 2 of 8 Claude Opus 4.5 (high reasoning, tau2) 5 May 2026
tau2 v1.0.1, airline: task success (pass^1)% of tasks, higher is better 84.0% 1 of 8 Claude Opus 4.5 (high reasoning, tau2) 5 May 2026
tau2 v1.0.1, banking knowledge, alltools retrieval: consistency (pass^4)% of tasks, higher is better 11.3% 15 of 21 Claude Opus 4.5 (high reasoning, tau2) 5 May 2026
tau2 v1.0.1, banking knowledge, alltools retrieval: task success (pass^1)% of tasks, higher is better 24.7% 17 of 21 Claude Opus 4.5 (high reasoning, tau2) 5 May 2026
tau2 v1.0.1, retail: consistency (pass^4)% of tasks, higher is better 51.8% 2 of 8 Claude Opus 4.5 (high reasoning, tau2) 5 May 2026
tau2 v1.0.1, retail: task success (pass^1)% of tasks, higher is better 79.6% 3 of 8 Claude Opus 4.5 (high reasoning, tau2) 5 May 2026
tau2 v1.0.1, telecom: consistency (pass^4)% of tasks, higher is better 78.1% 2 of 8 Claude Opus 4.5 (high reasoning, tau2) 5 May 2026
tau2 v1.0.1, telecom: task success (pass^1)% of tasks, higher is better 92.3% 2 of 8 Claude Opus 4.5 (high reasoning, tau2) 5 May 2026

Not shown: Not your policies or customers: the tasks and policies are tau2's own, and the customer is simulated by gpt-5.2.

UGI Leaderboard · reported by UGI Leaderboard · 6 results
MeasureValueRankConfigurationDated
UGI: requested-length error% off the requested word count, lower is better 3.0% 22 of 379 Claude Opus 4.5 (no reasoning) 24 Nov 2025
UGI: requested-length error% off the requested word count, lower is better 4.0% 32 of 379 Claude Opus 4.5 (thinking reasoning) 24 Nov 2025
UGI: style adherencescore from 0 to 1, higher is better 0.37 116 of 379 Claude Opus 4.5 (no reasoning) 24 Nov 2025
UGI: style adherencescore from 0 to 1, higher is better 0.39 41 of 379 Claude Opus 4.5 (thinking reasoning) 24 Nov 2025
UGI: writing scorescore out of 100, higher is better 70.4 19 of 379 Claude Opus 4.5 (no reasoning) 24 Nov 2025
UGI: writing scorescore out of 100, higher is better 70.3 20 of 379 Claude Opus 4.5 (thinking reasoning) 24 Nov 2025

Not shown: Not other format limits such as character counts or bullet counts.

Vectara Hallucination Leaderboard · reported by Vectara Hallucination Leaderboard · 2 results
MeasureValueRankConfigurationDated
Vectara: answer rate% of documents, higher is better 98.7% 74 of 108 Claude Opus 4.5 22 Sep 2026
Vectara: hallucination rate% of summaries, lower is better 10.9% 70 of 108 Claude Opus 4.5 22 Sep 2026

Not shown: Not a quality score: a low rate usually means content filters were triggered, and hallucination rates are measured on answered documents only.

Vending-Bench 2 · reported by Vending-Bench 2 · 1 result
MeasureValueRankConfigurationDated
Vending-Bench 2: money after a yearUS dollars, higher is better $4,967 29 of 63 Claude Opus 4.5 1 Oct 2026

Not shown: Not a real business; one simulated market with set rules.

Will Claude Opus 4.5 work for your product?

Benchmarks aren't your task. Paste the prompt your product uses and run it beside the models you could use instead, with every answer checked and costed.

New here? Your email starts a free trial. No credit card needed.