How much does Claude Opus 4.8 cost?

Claude Opus 4.8 costs $5.00/M input tokens and $25.00/M output tokens, with separate rates for Cache Read at $0.50/M tokens, Cache Write at $6.25/M tokens, Cache Write (1h) at $10.00/M tokens, and Web Search at $10.00/1K calls.

What is the context length of Claude Opus 4.8?

Claude Opus 4.8 has a 1,000,000 token context window. It supports up to 128,000 completion tokens.

Does Claude Opus 4.8 support tool calling and structured outputs?

Yes. Claude Opus 4.8 accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

What inputs and outputs does Claude Opus 4.8 support?

Claude Opus 4.8 accepts text, images, and files such as PDFs as input and returns text.

Which providers serve Claude Opus 4.8?

Claude Opus 4.8 is served by 5 providers on OpenRouter: Claude Platform on AWS, Azure, Amazon Bedrock, Google Vertex, and Anthropic. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.

When was Claude Opus 4.8 released?

Claude Opus 4.8 was released on May 27, 2026.

Anthropic: Claude Opus 4.8

anthropic/claude-opus-4.8

Compare

Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family. It supports text, image, and file inputs with text output, with reasoning support and a 1M-token context window. It is suited for highly autonomous agents, long-horizon agentic work, knowledge work, and memory-driven tasks where coherence over extended sessions matters.

It is particularly strong on multi-step reasoning, complex coding, and end-to-end project orchestration - large codebases, multi-stage debugging, and long-running asynchronous agent pipelines. Beyond coding, it handles knowledge work such as drafting documents, building presentations, and analyzing data, maintaining quality across very long outputs.

Modalities

In / out price

$5 / $25per 1M

Context

1.0M

Released

May 27, 2026

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick: Balanced (price + speed), Nitro (fastest), or Floor (cheapest).

Latency / throughput


	$5.00	$25.00	$0.50	0.72s	63 tok/s	99.91%
	$5.00	$25.00	$0.50	2.46s	65 tok/s	100.00%
	$5.00	$25.00	$0.50	0.89s	82 tok/s	99.72%
	$5.00	$25.00	$0.50	1.58s	71 tok/s	100.00%
	$5.00	$25.00	$0.50	0.98s	34 tok/s	98.82%
Not used in Standard routing:Why these endpoints are not used
	$5.50	$27.50	$0.55	--	--	--
	$5.50	$27.50	$0.55	0.53s	72 tok/s	100.00%
	$5.50	$27.50	$0.55	--	--	100.00%
	$5.50	$27.50	$0.55	4.08s	85 tok/s	100.00%
	$5.50	$27.50	$0.55	2.66s	77 tok/s	--
Fast	$10.00	$50.00	$1.00	0.92s	186 tok/s	100.00%

Input $5.00Output $25.00
63 tok/s99.91% uptime
Input $5.00Output $25.00
65 tok/s100.00% uptime
Input $5.00Output $25.00
82 tok/s99.72% uptime
Input $5.00Output $25.00
71 tok/s100.00% uptime
Input $5.00Output $25.00
34 tok/s98.82% uptime
Not used in Standard routing:Why these endpoints are not used
Input $5.50Output $27.50
-- tok/s-- uptime
Input $5.50Output $27.50
72 tok/s100.00% uptime
Input $5.50Output $27.50
-- tok/s100.00% uptime
Input $5.50Output $27.50
85 tok/s100.00% uptime
Input $5.50Output $27.50
77 tok/s-- uptime
Input $10.00Output $50.00
Fast
186 tok/s100.00% uptime

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Weighted average

Input

$2.544/M tokens

Output

$25.52/M tokens

Claude Opus 4.8 — Price history

Price history

No effective price data for this selection

		$0.916	$25.00	92.2%	19.6%
		$1.758	$25.00	74.7%	31.5%
		$3.164	$25.00	42.6%	10.9%
		$4.123	$25.00	55.3%	33.5%
		$5.000	$25.00	0.0%	0.0%
Not used in Standard routing:Why these endpoints are not used
	Fast	$1.879	$50.00	92.3%	4.3%
		$2.951	$27.50	62.0%	0.1%
		$2.965	$27.50	61.8%	0.0%
		$5.500	$27.50	0.0%	0.0%
		$5.896	$27.50	0.0%	0.0%
		—	—	—	—

Effective in $0.916Effective out $25.00
Token share
19.6%
Effective in $1.758Effective out $25.00
Token share
31.5%
Effective in $3.164Effective out $25.00
Token share
10.9%
Effective in $4.123Effective out $25.00
Token share
33.5%
Effective in $5.000Effective out $25.00
Token share
0.0%
Not used in Standard routing:Why these endpoints are not used
Effective in $1.879Effective out $50.00
FastToken share
4.3%
Effective in $2.951Effective out $27.50
Token share
0.1%
Effective in $2.965Effective out $27.50
Token share
0.0%
Effective in $5.500Effective out $27.50
Token share
0.0%
Effective in $5.896Effective out $27.50
Token share
0.0%
Effective in —Effective out —
Token share —

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Throughput

186tok/s

P50, best across providers

Latency

0.53s

P50, best provider

Reasoning effortLatency / throughput

AutoExacto benchmarks

GPQA DiamondTAU-BenchGoogle Vertex (US)89.9%78.2%Amazon Bedrock89.9%77.6%Google Vertex (EU)89.6%77.1%Anthropic90.0%76.3%Azure90.6%75.5%Google Vertex89.7%76.3%Amazon Bedrock (EU)90.5%75.4%Azure (US)89.9%76.0%auto-routing89.5%76.3%Amazon Bedrock (US)89.5%75.5%

+1 more providers

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Uptime (3d)

100.00%

Availability (3d)

99.84%

Availability over the last 3 days

Last 72 hours

Availability 99.84%

3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability

99.50%

Without Routing

97.92%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

Benchmarks

Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.

Benchmark score summary for Anthropic: Claude Opus 4.8 (Artificial Analysis, Design Arena and Vals AI)
Source	Benchmark	Score
Artificial Analysis	Claude Opus 4.8 (Max) Coding Index	74.3
Artificial Analysis	Claude Opus 4.8 (Max) Agentic Index	41.9
Artificial Analysis	Claude Opus 4.8 (Max) GPQA Diamond	92.0%
Artificial Analysis	Claude Opus 4.8 (Max) HLE	48.7%
Artificial Analysis	Claude Opus 4.8 (Max) IFBench	62.2%
Artificial Analysis	Claude Opus 4.8 (Max) τ²-Bench Telecom	94.4%
Artificial Analysis	Claude Opus 4.8 (Max) AA-LCR	77.7%
Artificial Analysis	Claude Opus 4.8 (Max) τ-Bench Banking	34.2%
Artificial Analysis	Claude Opus 4.8 (Max) GDPval-AA	48.1%
Artificial Analysis	Claude Opus 4.8 (Max) CritPt	20.9%
Artificial Analysis	Claude Opus 4.8 (Max) SciCode	54.4%
Artificial Analysis	Claude Opus 4.8 (Max) Terminal-Bench Hard	58.3%
Artificial Analysis	Claude Opus 4.8 (Max) Terminal-Bench 4.0	21.7%
Artificial Analysis	Claude Opus 4.8 (Max) Terminal-Bench 2.1	84.6%
Design Arena	Claude Opus 4.8 Agents Arena Agenticgamedev Elo	1185
Design Arena	Claude Opus 4.8 Agents Arena Agentichtmlslides Elo	1227
Design Arena	Claude Opus 4.8 Agents Arena Agenticslides Elo	1294
Design Arena	Claude Opus 4.8 Agents Arena Agenticslides(Html) Elo	1230
Design Arena	Claude Opus 4.8 Agents Arena Agenticslides(Python-Pptx) Elo	1310
Design Arena	Claude Opus 4.8 Agents Arena Androidnative Elo	1222
Design Arena	Claude Opus 4.8 Agents Arena Blender3d Elo	1096
Design Arena	Claude Opus 4.8 Agents Arena Codetovideo Elo	1187
Design Arena	Claude Opus 4.8 Agents Arena Full Stack Elo	1224
Design Arena	Claude Opus 4.8 Agents Arena Godotgamedev Elo	1211
Design Arena	Claude Opus 4.8 Agents Arena Htmlslides Elo	1116
Design Arena	Claude Opus 4.8 Agents Arena Mobile Apps Elo	1208
Design Arena	Claude Opus 4.8 Agents Arena Pdf Elo	1162
Design Arena	Claude Opus 4.8 Agents Arena Pptxslides Elo	1306
Design Arena	Claude Opus 4.8 Agents Arena Python-Pptxslides Elo	1298
Design Arena	Claude Opus 4.8 Agents Arena Python-Pptxslides Elo	1191
Design Arena	Claude Opus 4.8 Agents Arena Spreadsheet Elo	1018
Design Arena	Claude Opus 4.8 Agents Arena Webapps Elo	1214
Design Arena	Claude Opus 4.8 Models Arena 3D Elo	1201
Design Arena	Claude Opus 4.8 Models Arena Asciiart Elo	1270
Design Arena	Claude Opus 4.8 Models Arena Code Categories Elo	1243
Design Arena	Claude Opus 4.8 Models Arena Data Visualization Elo	1234
Design Arena	Claude Opus 4.8 Models Arena Game Development Elo	1250
Design Arena	Claude Opus 4.8 Models Arena SVG Elo	1182
Design Arena	Claude Opus 4.8 Models Arena UI Component Elo	1248
Design Arena	Claude Opus 4.8 Models Arena Website Elo	1250
Vals AI	Claude Opus 4.8 Code Migration	47.3%
Vals AI	Claude Opus 4.8 CorpFin v2	66.7%
Vals AI	Claude Opus 4.8 EMB	69.4%
Vals AI	Claude Opus 4.8 Finance Agent (v2)	53.9%
Vals AI	Claude Opus 4.8 GPQA Diamond	92.4%
Vals AI	Claude Opus 4.8 Harvey's Legal Agent Benchmark v1	9.6%
Vals AI	Claude Opus 4.8 LiveCodeBench	87.8%
Vals AI	Claude Opus 4.8 LegalBench	83.6%
Vals AI	Claude Opus 4.8 Legal Research Bench	43.8%
Vals AI	Claude Opus 4.8 MedCode	53.2%
Vals AI	Claude Opus 4.8 MedScribe	85.8%
Vals AI	Claude Opus 4.8 MMLU Pro	89.6%
Vals AI	Claude Opus 4.8 MMMU Pro	86.6%
Vals AI	Claude Opus 4.8 MortgageTax	69.9%
Vals AI	Claude Opus 4.8 ProgramBench	1.0%
Vals AI	Claude Opus 4.8 ProofBench v1	69.0%
Vals AI	Claude Opus 4.8 Public Benefits Bench v1.1	68.1%
Vals AI	Claude Opus 4.8 Vals RSI Index	24.0%
Vals AI	Claude Opus 4.8 SAGE v1	54.8%
Vals AI	Claude Opus 4.8 SkillsBench	59.2%
Vals AI	Claude Opus 4.8 SWE-bench	88.6%
Vals AI	Claude Opus 4.8 Tax Agent Bench	30.0%
Vals AI	Claude Opus 4.8 TaxEval v2	75.6%
Vals AI	Claude Opus 4.8 Terminal-Bench 4.0	23.2%
Vals AI	Claude Opus 4.8 Terminal-Bench-Science	4.3%
Vals AI	Claude Opus 4.8 Vals Index	55.1%
Vals AI	Claude Opus 4.8 Vals Multimodal Index	70.9%
Vals AI	Claude Opus 4.8 Vibe Code Bench v1.1	82.7%

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

Claude Code

Claude Code is Anthropic's agentic coding tool that reads your entire codebase, plans and executes changes across files, runs tests, and iterates on failures, all from natural language prompts.

212Btokens

Hermes Agent

Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.

34.7Btokens

OpenClaw

OpenClaw is an open-source AI agent that connects to your messaging apps and takes real actions on your behalf, from running commands and browsing the web to managing files and sending emails.

15.7Btokens

Cline

Cline is an open-source AI coding agent that lives inside your IDE, autonomously exploring your codebase, editing files, running terminal commands, and using browser automation.

8.08Btokens

There are many coding agents, but this one is yours.

6.95Btokens

Rest of OpenRouter

Top 5 Apps

1.67B

Anthropic: Claude Opus versions

Available through EU in-region routing

Anthropic: Claude Opus 4.8 is available through EU in-region routing. Requests sent to eu.openrouter.ai are decrypted in the EU and routed only to shared provider endpoints in Europe (the EU/EEA, the UK, and Switzerland).

Providers in Europe

Amazon Bedrock, Google Vertex

Prices in Europe

Amazon Bedrock: $5.50/M tokens input, $27.50/M tokens output
Google Vertex: $5.50/M tokens input, $27.50/M tokens output

Base URL

https://eu.openrouter.ai/api/v1

Plan

Requires Business or Enterprise. Self-serve on Business: no contract, no monthly minimum. OpenRouter Business for EU in-region routing

Fails closed: if no shared provider endpoint in Europe is available, the request errors instead of leaving the region. Requests served by your own BYOK key go to your deployment’s region, which you control. In-region routing docs

More models from Anthropic

Claude Fable 5.2

Claude Fable 5.2 is Anthropic's most capable model for the most demanding technical work, with gains over Fable 5.1 in coding, reasoning, and vision. In prerelease testing it completed otherwise-infeasible software migrations, designed working electronic devices from scratch, and made progress on open problems in physics and mathematics. It communicates more clearly than earlier Fable models and generates output up to 30% faster than Fable 5.1. Adaptive extended thinking is always on, with reasoning effort levels from low to max (default high) and preserved thinking across turns. Anthropic recommends Opus 5.5 as the daily driver for most coding and knowledge work, with Fable 5.2 for the most ambitious, open-ended problems.

Text1M context$8 / $40

Claude Fable 5.2 (batch)

Text1M context$4 / $20

Claude Haiku 5.5

Claude Haiku 5.5 is Anthropic's small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use. It succeeds Claude Haiku 4.5 with stronger coding, computer use, and knowledge work, and accepts text and image input with a 1M-token context window.

It is the first Haiku model with adjustable effort. Thinking is adaptive and on by default, with effort as the main lever for trading off depth, latency, and cost; it can be turned off at low, medium, and high effort.

Text1M context$0.10 / $0.50

Claude Haiku 5.5 (batch)

Text1M context$0.05 / $0.25

Claude Sonnet 5.5

Claude Sonnet 5.5 is Anthropic's Sonnet-class model for well-scoped everyday work, succeeding Claude Sonnet 5 as a direct upgrade. It is especially strong at building features, fixing bugs, and producing polished documents, slides, and spreadsheets, and it writes and communicates more clearly than its predecessor.

Thinking is always on, so effort is the main lever for trading off depth, latency, and cost, and lower effort settings keep it responsive for everyday agentic loops.

Text1M context$2 / $10

Claude Sonnet 5.5 (batch)

Thinking is always on, so effort is the main lever for trading off depth, latency, and cost, and lower effort settings keep it responsive for everyday agentic loops.

Text1M context$1 / $5

Claude Opus 5.5

Claude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, succeeding Claude Opus 5. It is particularly strong at multi-step changes in large codebases, code review and bug finding, financial and scientific analysis, and reading dense charts, diagrams, and screenshots, and it is more careful than its predecessor about only stating figures and citing sources it can back up.

The model completes comparable tasks in fewer steps and with fewer tokens than Opus 5, and reports on its work in plainer language, with clear updates on what it did, what it found, and what it needs from the user. Thinking is always adaptive, so effort is the main lever for trading off depth, latency, and cost, and lower effort settings remain effective for latency-sensitive workloads.

Text1M context$4 / $20

Claude Opus 5.5 (batch)

Text1M context$2 / $10

Claude Fable 5.1

Claude Fable 5.1 improves on Claude Fable 5 across the board, with the biggest gains in agentic coding, long-running agentic workflows, and knowledge work: long code refactors, front-end and visual code generation, and finance and analysis tasks in particular. It also tends to be more concise than Fable 5 in its plans and summaries. We recommend testing it as a direct upgrade wherever you use Fable 5 today, and alongside Opus 5 on reasoning-heavy tasks.

Text1M context$10 / $50

Claude Fable 5.1 (batch)

Text1M context$5 / $25

Claude Opus 5

Fast-mode variant of Opus 5 - identical capabilities with higher output speed at 2x pricing relative to regular Opus 5.

Learn more in Anthropic's docs: https://platform.claude.com/docs/en/build-with-claude/fast-mode

Note: As of September 1, 2026, this dedicated fast model is deprecated. Fast mode is now served by the fast service tier endpoint on Claude Opus 5 — request it with service_tier: "fast" or speed: "fast". Requests to this model keep working and are served by the same fast tier capacity, so no action is required, but new integrations should target the regular model. Learn more in our service tier docs: https://openrouter.ai/docs/guides/features/service-tiers

Text1M context

Claude Opus 5

Claude Opus 5 is Anthropic’s flagship model for demanding reasoning, coding, and long-horizon agentic work. It is particularly strong at end-to-end software tasks, code review and bug finding, visual analysis of charts and documents, complex office deliverables, and coordinating parallel subagents.

The model maintains strong instruction following and tool use across extended tasks, while remaining effective at lower effort settings for workloads that prioritize latency and token efficiency.

Text1M context$5 / $25

Claude Opus 5 (batch)

The model maintains strong instruction following and tool use across extended tasks, while remaining effective at lower effort settings for workloads that prioritize latency and token efficiency.

Text1M context$2.50 / $12.50

Claude Sonnet 5

Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max, and x-high), a 1M-token context window, and text, image, and file inputs. Sonnet 5 uses an updated tokenizer and includes real-time cyber safeguards that block certain high-risk dual-use activities.

Text1M context$2 / $10

Claude Sonnet 5 (batch)

Text1M context$1 / $5

Claude Fable Latest

This model always redirects to the latest model in the Claude Fable family.

Text1M context

Claude Fable 5

Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding. It supports text, image, and file inputs with text output, with reasoning support and a 1M-token context window. It is suited for long-running, complex, and asynchronous tasks that previously required frequent human check-ins.

It is particularly strong at end-to-end work that would otherwise take a person hours, days, or weeks - taking on problems that are long-running, ambiguous, or highly multi-step. It executes well-scoped tasks with few mistakes, automatically self-correcting through verification loops, and ships with robust safeguards.

Text1M context$10 / $50

Claude Fable 5 (batch)

Text1M context$5 / $25

Claude Opus 4.8

Fast-mode variant of Opus 4.8 - identical capabilities with higher output speed at 2x pricing relative to regular Opus 4.8.

Learn more in Anthropic's docs: https://platform.claude.com/docs/en/build-with-claude/fast-mode

Note: As of September 1, 2026, this dedicated fast model is deprecated. Fast mode is now served by the fast service tier endpoint on Claude Opus 4.8 — request it with service_tier: "fast" or speed: "fast". Requests to this model keep working and are served by the same fast tier capacity, so no action is required, but new integrations should target the regular model. Learn more in our service tier docs: https://openrouter.ai/docs/guides/features/service-tiers

Text1M context

Claude Opus 4.8

Text1M context$5 / $25

Anthropic: Claude Opus 4.8

Providers