Models · Anthropic · Out sinceReleased 24 Nov 2025

Claude Opus 4.5#23 for writing.#23 for writing, best at high effort.

Available on OpenRouterOn OpenRouterAvailable nowGenerally availableClosed (can't be downloaded)Proprietary

Claude Opus 4.5 is made by Anthropic. Among the models we track it ranks #23 for writing. It's expensive to use.

Auto-created from OpenRouter catalog; verify details.

Writing & creativity: 50.3 / 100 · #23
Research & analysis: —
Coding: —
Price: Expensive$5 / $25
Price per 1M (blended)Blended / 1M: $10
MemoryContext: 200K
Longest answerMax output: —
Test resultsResults: 15 (15 independent15 indep.)
Out sinceReleased: 24 Nov 2025
Made byVendor: Anthropic
UnderstandsInputs: —

Scores are out of 100 for each area, compared with every model we track; “#” is its rank. Context is how much text it can read at once (1M is roughly 700,000 words). Blended price mixes the cost of what you send and what it writes back.

Thinking levels

Reasoning settings

You can't choose how long this one thinks.

No adjustable reasoning setting listed.

Links

Using it through OpenRouter

On OpenRouter

OpenRouter is a service that gives access to many AI models in one place. This model is listed there as anthropic/claude-opus-4.5.

Route it as anthropic/claude-opus-4.5 at $5 in / $25 out per 1M tokens, 200K context. Listed since 24 Nov 2025.

include_reasoningmax_completion_tokensmax_tokensreasoningresponse_formatstopstructured_outputstemperaturetool_choicetoolstop_kverbosity

Every test result

Every result

All the numbers,with where they came from.

Every result we've found for this model, with who measured it and a link to where we read it.

All raw rows (vendor and independent kept separate), with setting label, source, date, harness and notes.

All results for this model
						NotesNotes
Humanity's Last Exam	25.2%	Defaultclaude-opus-4-5-20251101-thinking	Independent testIndependentScale AI SEAL ↗	1 Oct 2026	—	rank 10 (Scale rank accounts for CI); ±1.7; entry added 2025-11-26
LMArena Search Arena	1180	Defaultclaude-opus-4-5-search	Independent testIndependentLMArena ↗	24 Aug 2026	—	rank 19 (rank range 18-22); 95% CI 1174.0-1185.3; 61573 votes
LMArena Text - Coding category	1531	Highclaude-opus-4-5-20251101-high-32k	Independent testIndependentLMArena ↗	30 Sep 2026	—	text arena coding category, style control on; rank 19 (CI rank 7-42); 95% CI 1523-1538; 7790 votes
LMArena Text - Coding category	1523	Defaultclaude-opus-4-5-20251101	Independent testIndependentLMArena ↗	30 Sep 2026	—	text arena coding category, style control on; rank 27 (CI rank 8-53); 95% CI 1518-1528; 17972 votes
LMArena Text - Creative Writing	1469	Highclaude-opus-4-5-20251101-high-32k	Independent testIndependentLMArena ↗	30 Sep 2026	—	rank 18 (rank range 7-35); 95% CI 1460.8-1477.4; 5677 votes; style-controlled
LMArena Text - Instruction Following	1483	Highclaude-opus-4-5-20251101-high-32k	Independent testIndependentLMArena ↗	30 Sep 2026	—	rank 17 (rank range 8-35); 95% CI 1476.5-1489.8; 9867 votes; style-controlled
LMArena Text - Longer Query	1495	Highclaude-opus-4-5-20251101-high-32k	Independent testIndependentLMArena ↗	30 Sep 2026	—	rank 19 (rank range 9-37); 95% CI 1488.5-1501.7; 9830 votes; style-controlled
MCP Atlas	69.8%	Highclaude-opus-4-5 (high)	Independent testIndependentScale AI SEAL ↗	1 Oct 2026	—	rank 13 (Scale rank accounts for CI); ±2.9; entry added 2025-09-10
METR 50% time horizon	4.9 h	Default	Independent testIndependentMETR ↗	8 May 2026	METR react agent (Inspect)	50% time horizon, METR-Horizon-v1.1; 95% CI 162-624 min; 80% horizon 49.4 min; release 2025-11-24; raw data https://metr.org/assets/benchmark_results_1_1.yaml
PRBench Finance (Scale)	46.2%	Defaultclaude-opus-4-5-20251101-thinking	Independent testIndependentScale AI (SEAL) ↗	26 Nov 2025	—	Scale rank 13; ±0.27 CI
PRBench Legal (Scale)	44.2%	Defaultclaude-opus-4-5-20251101-thinking	Independent testIndependentScale AI (SEAL) ↗	26 Nov 2025	—	Scale rank 18; ±0.34 CI
SWE-bench Multilingual	70.7%	Default	Independent testIndependentSWE-bench ↗	13 Feb 2026	mini-SWE-agent 2.0.0a0	Official SWE-bench bash-only (mini-SWE-agent) run; data file https://raw.githubusercontent.com/SWE-bench/swe-bench.github.io/master/data/leaderboards.json; leaderboard not updated since 2026-02-26
SWE-Bench Pro (public, v1)	45.9%	Defaultclaude-opus-4-5-20251101	Independent testIndependentScale AI SEAL ↗	1 Oct 2026	—	rank 5 (Scale rank accounts for CI); ±3.6; entry added 2025-12-11
SWE-bench Verified (bash-only, mini-SWE-agent)	76.8%	High	Independent testIndependentSWE-bench ↗	17 Feb 2026	mini-SWE-agent 2.0.0	Official SWE-bench bash-only (mini-SWE-agent) run; data file https://raw.githubusercontent.com/SWE-bench/swe-bench.github.io/master/data/leaderboards.json; leaderboard not updated since 2026-02-26
Vectara Hallucination Leaderboard (HHEM)	10.9%	Defaultanthropic/claude-opus-4-5-20251101	Independent testIndependentVectara ↗	22 Sep 2026	—	factual consistency 89.1 %; answer rate 98.7 %; avg summary 114.5 words; HHEM-2.3 judge; effort not stated (API default)