Claude Opus 4.7: Benchmarks, Pricing & Context | hokai.io

Claude Opus 4.7 by Anthropic (April 2026): 87.6% SWE-bench, 1M token context, 3.75MP vision. $5/$25 per 1M tokens. Top GA model for agentic coding tasks.

Claude Opus 4.7, released April 16, 2026 by Anthropic, is the top generally available model for agentic coding with 87.6% on SWE-bench Verified. It supports a 1M token context window and 3.75MP vision. Pricing is $5 per million input tokens and $25 per million output tokens. Adaptive thinking replaces extended thinking budgets. Output speed is 42.2 tokens per second, making it unsuitable for real-time applications.

Claude Opus 4.7 is Anthropic's most capable generally available model, released April 16, 2026. It scores 87.6% on SWE-bench Verified and 94.2% on GPQA Diamond. The context window is 1 million tokens with 128K max output. Pricing is $5 per million input tokens and $25 per million output tokens. It leads all GA models on tool use (MCP-Atlas 77.3%) and is the first Claude with 3.75MP high-resolution vision.

Provider: Anthropic · Family: Claude 4

Context window: 1,000,000 tokens · Max output: 128,000

Input modalities: text, image, pdf, tool-calls · Output: text, tool-calls

About Claude Opus 4.7

Claude Opus 4.7 is Anthropic's most capable generally available model, released on April 16, 2026. Built on a dense Transformer architecture (not MoE) with undisclosed parameter count, it is the fourth major Opus iteration in the Claude 4 generation, following Opus 4.1, 4.5, 4.6, and now 4.7. Anthropic positioned it specifically to close the gap between their generally-available lineup and frontier research previews, with a deliberate focus on agentic coding, long-horizon autonomous tasks, and high-fidelity vision understanding. It replaces Claude Opus 4.6 as Anthropic's recommended starting point for any complex workload. On benchmarks, Claude Opus 4.7 posts 87.6% on SWE-bench Verified, a 6.8-percentage-point gain over Opus 4.6's 80.8% and ahead of GPT-5.4 on the same leaderboard. SWE-bench Pro, a harder multi-file variant, jumped from 53.4% on Opus 4.6 to 64.3% on Opus 4.7, leapfrogging GPT-5.4 (57.7%) and Gemini 3.1 Pro (54.2%). On GPQA Diamond, which tests PhD-level scientific reasoning across physics, chemistry, and biology, Opus 4.7 scores 94.2%, effectively tied with GPT-5.4 at 94.4% and Gemini 3.1 Pro at 94.3% — that benchmark is approaching saturation. For tool use, MCP-Atlas scores 77.3%, ahead of Opus 4.6 at 75.8%, GPT-5.4 at 68.1%, and Gemini 3.1 Pro at 73.9%. Computer use on OSWorld-Verified reached 78.0%, ahead of GPT-5.4 at 75.0%. Humanity's Last Exam without tools sits at 46.9%, and with tools at 54.7%. The context window is 1,000,000 tokens — the full 1M provided at standard per-token rates with no long-context premium. Maximum output via the synchronous Messages API is 128,000 tokens. On the Batch API with the output-300k-2026-03-24 beta header, Opus 4.7 can produce up to 300,000 output tokens per request. A single request can include up to 600 images or PDF pages. Note: Opus 4.7 uses a new tokenizer that processes text with up to 35% more tokens than Opus 4.6 for equivalent input, so effective costs on existing prompts can increase even though the rate card is unchanged. Modalities supported as inputs are text and images (PDF documents are parsed as images). Output is text only. No native audio input or output is available. Opus 4.7 is the first Claude model with high-resolution image support, raising maximum image resolution from 1568px / 1.15MP to 2576px / 3.75MP. This is more than a 3x increase in pixel area, which meaningfully improves performance on dense UIs, technical diagrams, chemical structure reading, and computer use screenshots. Image coordinates now map 1:1 to actual pixels, removing the need for scale-factor math in agentic loops. Tool use and function calling are fully supported with native parallel tool calls, client-side tool schemas, and server-side tools including web search, web fetch, and code execution. Computer use is available via the beta API. Task budgets (beta) allow the model to self-regulate token spend across a full agentic loop. The new xhigh effort level sits between high and max, giving finer control over reasoning depth versus speed. Pricing is $5.00 per 1M input tokens and $25.00 per 1M output tokens, unchanged from Opus 4.6. Prompt caching cuts cached input reads to $0.50 per 1M (10% of base rate); a 5-minute cache write costs $6.25 per 1M and a 1-hour cache write costs $10.00 per 1M. Batch API processing cuts both input and output by 50%, to $2.50 and $12.50 per 1M respectively. US-only inference via the inference_geo parameter adds a 1.1x multiplier. For a 100,000-token document summarization task (100K input, 1K output), the uncached cost is $0.525. A daily coding agent running 1M input tokens and 200K output tokens costs $6.00. Customer support at 2,000 tokens in and 500 tokens out across 1,000 turns costs $13.50 before caching discounts. Claude Opus 4.7 is available via the Anthropic API (model ID: claude-opus-4-7), AWS Bedrock (ID: anthropic.claude-opus-4-7-v1:0), Google Cloud Vertex AI (ID: claude-opus-4-7@20260416), and Microsoft Foundry (via the Azure AI Foundry catalog). Authentication is via API key on Anthropic direct, AWS IAM on Bedrock, GCP IAM on Vertex, and Azure subscription billing on Foundry. Microsoft Foundry regions are restricted to East US2 and Sweden Central. AWS Bedrock supports both global endpoints (dynamic routing) and regional endpoints (geo-guaranteed routing). Cloud providers add a margin on top of Anthropic's base rates for enterprise SLAs and region pinning. Safety evaluation for Opus 4.7 was published in the system card released April 16, 2026. The model uses Constitutional AI and RLHF alignment. In multi-turn safety evaluations, Opus 4.7 successfully identified escalating harm patterns even when individual prompts appeared benign. In agentic settings, it is better than Opus 4.6 at refusing malicious instructions and resisting prompt injection in Claude Code and computer use contexts. Cyber capabilities are roughly similar to Opus 4.6; an external evaluation by the UK AI Security Institute found it could not complete their full cyber range, unlike the more capable Mythos Preview. Evaluation awareness was detected in about 9% of transcripts, triggered by inconsistencies in mocked tool results. Cybersecurity refusals are new real-time safeguards; teams doing legitimate security work can apply to the Cyber Verification Program. Claude Opus 4.7 is the right model for teams running autonomous coding agents over long multi-file sessions, where its 87.6% SWE-bench Verified score and 77.3% MCP-Atlas tool-use rating provide a measurable edge. It is also strong for enterprise knowledge work requiring long-document analysis — the 1M context at standard pricing means no special endpoint is needed to ingest a full codebase or document corpus. The model excels at vision-heavy agentic tasks like UI automation and chart interpretation, given the 3.3x resolution upgrade and 1:1 pixel coordinate mapping. Teams should not choose Opus 4.7 for latency-sensitive voice or chat applications: time to first token is approximately 19 seconds and output speed is 42.2 tokens per second, notably below the frontier median of 59.8 t/s. For cost-sensitive high-volume tasks, Claude Sonnet 4.6 at $3/$15 per 1M is a strong alternative that maintains near-Opus coding performance at lower latency. Training data cutoff is January 2026, meaning the model has reliable knowledge of events through that date. Anthropic does not train on API inputs by default. Enterprise zero-retention plans are available. The model is SOC 2 Type II compliant, HIPAA-eligible, ISO 27001 certified, and GDPR compliant. EU AI Act classification is general-purpose AI with systemic risk obligations. API inputs and outputs are retained for 30 days for abuse monitoring and deleted unless flagged, unless the enterprise zero-retention option is enabled. Versus Opus 4.6, the headline gains are: SWE-bench Verified up 6.8pp to 87.6%, SWE-bench Pro up 10.9pp to 64.3%, OSWorld-Verified computer use up from 72.7% to 78.0%, and image resolution up 3.3x to 3.75MP. Breaking changes shipped with Opus 4.7 include removal of extended thinking budgets (adaptive thinking is now the only mode), removal of sampling parameter overrides (temperature, top_p, top_k now return 400 errors), and a new tokenizer that increases token counts by up to 35%. Some developers have reported regressions on specific workflows, particularly those relying on precise temperature control or verbose chain-of-thought output.

Pricing

$5.00 per 1M input tokens, $25.00 per 1M output tokens. Prompt caching: 5-min cache write at $6.25/MTok, 1-hour cache write at $10.00/MTok, cache reads at $0.50/MTok (90% savings). Batch API: 50% discount at $2.50 input / $12.50 output per MTok. US-only inference adds 1.1x multiplier. New tokenizer may increase effective token counts by up to 35% vs Opus 4.6.

Key Features

• High-Resolution Vision (3.75MP): First Claude model to support 2576px / 3.75MP images, a 3.3x increase over previous models. Pixel coordinates map 1:1 to actual image pixels, removing scale-factor math from computer use agents.
• 1M Token Context at Standard Pricing: Full 1 million token context window with no long-context premium. Up to 600 images or PDF pages per request. Reliable knowledge through January 2026.
• Adaptive Thinking with xhigh Effort: New xhigh effort level between high and max for finer reasoning/latency control. Adaptive thinking (replacing extended thinking budgets) is the only reasoning mode and outperforms explicit budget setting in internal evals.
• Task Budgets for Agentic Loops: Beta feature: give Claude an advisory token target across a full agentic session. Model self-regulates via a running countdown, finishing tasks gracefully without runaway token spend.
• Best-in-Class Tool Use: MCP-Atlas 77.3%, highest among generally available models. Supports native parallel tool calls, client-side schemas, and server-side tools including web search ($10/1K searches), web fetch (free), and code execution.
• Prompt Caching (90% Savings): Cache system prompts, tool definitions, and document context at $0.50/MTok (10% of base input rate). Stack with Batch API for up to 95% savings vs uncached standard requests.

Pros

• 87.6% SWE-bench Verified, the top score among generally available models as of April 2026.
• 1M context window at $5/MTok with no surcharge, enabling full codebase or document corpus ingestion in a single call.
• MCP-Atlas 77.3% tool use, the highest generally available score, critical for reliable multi-tool agent orchestration.

Cons

• Time to first token of ~19 seconds and output speed of 42.2 t/s, well below the 59.8 t/s frontier median, ruling it out for latency-sensitive applications.
• New tokenizer silently increases token counts by up to 35%, meaning existing prompts can cost more despite unchanged per-token rates.
• Breaking API changes at launch: temperature/top_p/top_k removed, extended thinking budget syntax removed — requires migration work for all Opus 4.6 integrations.

Benchmarks

• mmlu: 91.5
• gpqa diamond: 94.2
• swe bench verified: 87.6
• humanitys last exam: 46.9

Visit Claude Opus 4.7 Official Page