--- title: 'Claude Opus 4.7: Benchmarks, Features & Pricing' description: Claude Opus 4.7 leads SWE-bench Pro at 64.3% and OSWorld at 78.0%. Full breakdown of benchmarks, new features, pricing, and what changed from Claude Opus 4.6. url: https://nerdleveltech.com/claude-opus-4-7-benchmarks-features-pricing language: en date: '2026-04-17' category: ai-ml tags: - Claude - Anthropic - Claude Opus 4.7 - AI benchmarks - LLM - AI agents - coding AI - extended thinking - AI pricing markdown_url: https://nerdleveltech.com/claude-opus-4-7-benchmarks-features-pricing.md attribution: 'Source: NerdLevelTech — https://nerdleveltech.com/claude-opus-4-7-benchmarks-features-pricing' terms: https://nerdleveltech.com/llms.txt --- # Claude Opus 4.7: Benchmarks, Features & Pricing ## TL;DR Anthropic released Claude Opus 4.7 on April 16, 2026. It leads SWE-bench Pro at 64.3% — ahead of GPT-5.4 (57.7%) and Gemini 3.1 Pro (54.2%) — and sets a new high on OSWorld-Verified at 78.0%, up from 72.7% for Opus 4.6. Pricing is unchanged at $5.00/$25.00 per million tokens input/output, though a new tokenizer uses up to 35% more tokens for equivalent text. Key additions include a new `xhigh` effort level, adaptive thinking (replacing extended thinking budgets), task budgets for agentic loops, and 3× higher image resolution for computer use. --- ## What You'll Learn - How Claude Opus 4.7 scores on SWE-bench Pro, OSWorld, GPQA, CursorBench, and other key benchmarks - What changed vs. Claude Opus 4.6 — and what was removed - The new `xhigh` effort level and adaptive thinking system - Pricing, tokenizer changes, and what they mean for real API costs - Where Opus 4.7 leads, where competitors catch up, and what's still invitation-only --- ## Release Details Claude Opus 4.7 became generally available on April 16, 2026, one day before this post. The API model ID is `claude-opus-4-7`, available on Anthropic's API, Amazon Bedrock, Google Cloud Vertex AI, Microsoft Foundry, Snowflake Cortex AI, and GitHub Copilot Enterprise[^1]. The model ships as a single variant — there are no Thinking, Pro, or Mini tiers for Opus 4.7. Extended thinking budgets have been removed entirely (setting `budget_tokens` now returns a 400 error); adaptive thinking replaces them. Knowledge cutoff is January 2026[^2]. --- ## Benchmark Results ### Coding — Where Opus 4.7 Leads Most Clearly SWE-bench is the standard measure for autonomous software engineering: given a GitHub issue, can the model write a pull request that passes the test suite? Opus 4.7 sets a new top score on the harder **SWE-bench Pro** variant: | Model | SWE-bench Pro | SWE-bench Verified | |-------|--------------|-------------------| | Claude Opus 4.7 | **64.3%** | **87.6%** | | GPT-5.4 | 57.7% | — | | Gemini 3.1 Pro | 54.2% | 80.6% | | Claude Opus 4.6 | 53.4% | 80.8% | On SWE-bench Pro, Opus 4.7 leads by more than 6 points over GPT-5.4 and 10 points over Gemini. On the standard SWE-bench Verified leaderboard, Opus 4.7 climbs from 80.8% to 87.6% — a 6.8-point improvement over its predecessor[^3]. Anthropic also reports a **13% improvement** on an internal 93-task coding benchmark, and Rakuten's production deployment found Opus 4.7 resolves **3× more real production tasks** than Opus 4.6 on their SWE-bench variant[^4]. ### Computer Use — OSWorld OSWorld-Verified measures autonomous desktop task completion (file management, browser navigation, multi-app workflows). The human expert baseline is approximately 72.4%: | Model | OSWorld-Verified | |-------|-----------------| | GPT-5.4 | 75.0% | | Claude Opus 4.7 | **78.0%** | | Claude Opus 4.6 | 72.7% | | Human expert baseline | ~72.4% | Opus 4.7 moves past GPT-5.4 on this benchmark, reaching 78.0% and extending the gap above the human baseline. The improvement is partly driven by a new 3.75MP image resolution ceiling for computer use — up from 1.15MP in Opus 4.6 — and coordinates that map 1:1 with pixels, removing the scale-factor math that previously introduced errors in screen coordinate targeting[^5]. ### Developer Workflows — CursorBench CursorBench evaluates real-world coding assistant tasks as they occur in an IDE environment. Opus 4.7 scores **70%**, up from 58% for Opus 4.6 — a 12-point jump that positions it above competing models on this benchmark[^6]. These gains show up most in agentic tooling like [Claude Code's agent mode](https://nerdleveltech.com/claude-code-agent-mode-the-future-of-autonomous-coding), where the model reads, writes, and executes across a whole repository. ### Graduate-Level Science — GPQA Diamond On GPQA Diamond (graduate-level physics, chemistry, biology), the three frontier models are statistically indistinguishable: | Model | GPQA Diamond | |-------|-------------| | GPT-5.4 Pro | 94.4% | | Gemini 3.1 Pro | 94.3% | | Claude Opus 4.7 | **94.2%** | The differences here are within measurement noise. No single model holds a meaningful advantage on graduate-level scientific reasoning[^7]. ### Knowledge Work — GDPVal-AA GDPVal-AA is an Elo-based benchmark measuring general knowledge work across business analysis, document processing, and professional reasoning tasks: | Model | GDPVal-AA (Elo) | |-------|----------------| | Claude Opus 4.7 | **1,753** | | GPT-5.4 | 1,674 | | Gemini 3.1 Pro | 1,314 | Opus 4.7 holds a 79-point advantage over GPT-5.4 on this benchmark, with Gemini 3.1 Pro trailing significantly at 1,314[^8]. ### Security — XBOW Visual Acuity On XBOW's visual acuity cybersecurity benchmark, Opus 4.7 scores **98.5%**, versus 54.5% for Opus 4.6 — the largest single-generation jump of any benchmark in this release. Anthropic describes Opus 4.7 as a testbed for new cyber safeguards being validated before eventual broader release of Mythos-class models[^9]. ### Legal — BigLaw Bench On Harvey's BigLaw Bench (professional legal reasoning), Opus 4.7 scores **90.9%** at high effort[^10]. --- ## What's New vs. Opus 4.6 ### Adaptive Thinking Replaces Extended Thinking Budgets The biggest architectural change: extended thinking budgets are gone. Setting `budget_tokens` in your API request now returns a 400 error. In their place, Anthropic introduces **adaptive thinking** — off by default, opt-in via the API — which Anthropic says outperforms extended thinking in internal evaluations. The system dynamically allocates reasoning compute rather than requiring developers to set a token ceiling. Thinking blocks still appear in the response stream but the `thinking` field is empty by default unless you set `display: "summarized"` in your request[^11]. ### New `xhigh` Effort Level A new `xhigh` effort tier sits between the previous `high` and `max` levels, giving developers finer control over the reasoning/latency tradeoff. Anthropic recommends `xhigh` for coding and agentic use cases where you want stronger reasoning without paying the full `max` cost[^12]. ### Task Budgets for Agentic Loops A new public beta feature (`task-budgets-2026-03-13` beta header) lets you set an advisory token budget across an entire agentic loop — not just a single model call. The minimum is 20,000 tokens. This is not a hard cap but guides the model's planning toward completing the task within budget, reducing runaway token usage in long agent workflows[^13]. ### 3× Higher Image Resolution for Computer Use Max image resolution climbs from 1,568px (1.15MP) to **2,576px (3.75MP)** — more than three times the pixel count. Coordinate output now maps 1:1 to actual pixel positions, eliminating the scale-factor conversion errors that caused missed clicks in Opus 4.6 computer use deployments. For a deeper look at how Claude's computer use capabilities are reshaping agent workflows, see our post on [Claude managed agents](https://nerdleveltech.com/claude-managed-agents-deploy-ai-agents-faster). ### Sampling Parameters Removed Setting `temperature`, `top_p`, or `top_k` to non-default values now returns a 400 error. Anthropic has taken full control of sampling for Opus 4.7. Developers who relied on temperature tuning for output diversity will need to prompt-engineer instead. ### More Direct, Less Deferential Tone Opus 4.7 is described as more direct and opinionated than 4.6 — less validation-forward, with fewer emoji in responses and stronger opinions when asked. At lower effort levels, it is more literal and will not silently generalize instructions it considers ambiguous[^14]. ### `/ultrareview` in Claude Code A new `/ultrareview` slash command is available in Claude Code for deeper code review passes[^15]. --- ## Pricing and Real-World Cost The published rate card is unchanged from Opus 4.6: | Tier | Input | Output | |------|-------|--------| | Standard | $5.00 / MTok | $25.00 / MTok | | Batch API (50% off) | $2.50 / MTok | $12.50 / MTok | | Cache reads | $0.50 / MTok | — | | Cache writes (5-min) | $6.25 / MTok | — | | Cache writes (1-hour) | $10.00 / MTok | — | **The critical caveat:** Opus 4.7 uses a new tokenizer that converts the same input text into **up to 35% more tokens** than older Claude models. The per-token price is unchanged, but the effective per-request cost is higher. Developers migrating from Opus 4.6 should benchmark their actual token consumption before assuming cost parity[^16]. For reference, Gemini 3.1 Pro is available at approximately $2.00 input / $12.00 output per million tokens — roughly 2.5× cheaper at list price — though with a different capability profile and no published SWE-bench Pro or GDPVal-AA scores to compare directly. --- ## What Opus 4.7 Is Not: Claude Mythos Preview Anthropic's highest-capability model is not Opus 4.7. **Claude Mythos Preview** — developed under Project Glasswing — launched with 12 named enterprise and government partners, with access extended to over 40 additional vetted organizations on an invitation-only basis. Named launch partners are Anthropic, AWS, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorganChase, Linux Foundation, Microsoft, Nvidia, and Palo Alto Networks. Mythos is focused on defensive cybersecurity workflows. Anthropic describes it as "more capable and better-aligned" than Opus 4.7 in their internal evaluations. Opus 4.7 is explicitly positioned as the model on which Anthropic is testing new cyber safeguards — validating approaches before eventually moving toward a broader Mythos-class release. For context on Mythos's cybersecurity evaluation, see our post on the [AISI Claude Mythos cyber evaluation](https://nerdleveltech.com/aisi-claude-mythos-cyber-evaluation). --- ## Safety Profile Anthropic characterizes Opus 4.7's alignment as "largely well-aligned and trustworthy, though not fully ideal in its behavior." Specific improvements over 4.6 include better honesty and improved resistance to prompt injection attacks. One noted regression: Opus 4.7 is "modestly weaker" than 4.6 on avoiding overly detailed harm-reduction advice on controlled substances[^17]. The model automatically detects and blocks requests indicating prohibited or high-risk cybersecurity uses. A Cyber Verification Program is available for legitimate security professionals who need expanded capabilities. A full Claude Opus 4.7 System Card has been published[^18]. --- ## How It Compares: The Short Version Opus 4.7 is the strongest model for autonomous coding and agentic tasks right now. On SWE-bench Pro it leads by over 6 points; on OSWorld it overtakes GPT-5.4 at 78.0%. For general scientific reasoning (GPQA Diamond), the gap between frontier models has effectively collapsed — all three sit within 0.2 points of each other. Where Opus 4.7 loses ground: pricing (Gemini 3.1 Pro is cheaper at list price at approximately $2.00/$12.00 per million tokens, though benchmark comparisons are incomplete). Context windows are at parity — both models support 1M tokens. If you're optimizing for cost in long-document workflows rather than coding agents, the tradeoffs shift. For a broader look at where AI agents now stand relative to human performance across every benchmark category, see the [Stanford AI Index 2026](https://nerdleveltech.com/stanford-ai-index-report-us-china-gap-adoption) and our earlier breakdown of [GPT-5.4's computer use scores](https://nerdleveltech.com/gpt-5-4-beats-humans-computer-use-ai-agents). --- ## Frequently Asked Questions **What is Claude Opus 4.7?** Claude Opus 4.7 is Anthropic's latest publicly available large language model, released April 16, 2026. It is the most capable Claude model available to general users, with top scores on coding (SWE-bench Pro: 64.3%) and computer use (OSWorld-Verified: 78.0%) benchmarks. **What is the API model ID for Claude Opus 4.7?** The official model ID is `claude-opus-4-7`. It is available on Anthropic's API, Amazon Bedrock, Google Cloud Vertex AI, Microsoft Foundry, Snowflake Cortex AI, and GitHub Copilot Enterprise. **How much does Claude Opus 4.7 cost?** Standard pricing is $5.00 per million input tokens and $25.00 per million output tokens. Batch API pricing is $2.50/$12.50. Important: a new tokenizer uses up to 35% more tokens for the same text compared to older Claude models, so effective per-request costs are higher than the rate card alone suggests. **What happened to extended thinking budgets?** Extended thinking budgets have been removed in Opus 4.7. Setting `budget_tokens` returns a 400 error. Adaptive thinking — which dynamically allocates reasoning compute — replaces them and is off by default. Anthropic reports it outperforms extended thinking in internal evals. **How does Claude Opus 4.7 compare to GPT-5.4?** On SWE-bench Pro (real-world coding), Opus 4.7 leads 64.3% vs. 57.7%. On OSWorld computer use, Opus 4.7 leads 78.0% vs. 75.0%. On GPQA Diamond (science), they are within 0.2 points (94.2% vs. 94.4%). On GDPVal-AA knowledge work, Opus 4.7 leads 1,753 vs. 1,674 Elo. **Is Claude Opus 4.7 Anthropic's most powerful model?** No. Claude Mythos Preview (Project Glasswing) is more capable and better-aligned per Anthropic's internal evaluations, but remains invitation-only. It launched with 12 named partners — including AWS, Apple, Google, Microsoft, and Nvidia — with access extended to over 40 additional vetted organizations. Opus 4.7 is the most capable model available to the general public. --- ## References --- ## Related reads - [Cursor AI Editor 2.5 Review 2026: Models, Pricing, Verdict](https://nerdleveltech.com/cursor-ai-editor-the-2026-developers-power-tool) [^1]: [Claude Opus 4.7 announcement — Anthropic](https://www.anthropic.com/news/claude-opus-4-7) [^2]: [Claude models overview — Anthropic platform docs](https://platform.claude.com/docs/en/about-claude/models/overview) [^3]: [Claude Opus 4.7 SWE-bench Pro scores — The Next Web](https://thenextweb.com/news/anthropic-claude-opus-4-7-coding-agentic-benchmarks-release) [^4]: [Claude Opus 4.7 coding benchmarks — Anthropic announcement](https://www.anthropic.com/news/claude-opus-4-7) [^5]: [What's new in Claude 4.7 — Anthropic platform docs](https://platform.claude.com/docs/en/about-claude/models/whats-new-claude-4-7) [^6]: [CursorBench scores — Anthropic announcement](https://www.anthropic.com/news/claude-opus-4-7) [^7]: [GPQA Diamond benchmark comparison — Startup Fortune](https://startupfortune.com/anthropics-claude-opus-47-benchmarks-confirm-the-companys-most-capable-model-yet-and-pile-pressure-on-openai-and-google/) [^8]: [GDPVal-AA Elo scores — benchmark aggregators citing Anthropic data](https://www.anthropic.com/news/claude-opus-4-7) [^9]: [XBOW visual acuity benchmark and cyber safeguards — Anthropic announcement](https://www.anthropic.com/news/claude-opus-4-7) [^10]: [BigLaw Bench — Anthropic announcement](https://www.anthropic.com/news/claude-opus-4-7) [^11]: [Adaptive thinking replaces extended thinking budgets — Anthropic platform docs](https://platform.claude.com/docs/en/about-claude/models/whats-new-claude-4-7) [^12]: [xhigh effort level — Anthropic platform docs](https://platform.claude.com/docs/en/about-claude/models/whats-new-claude-4-7) [^13]: [Task budgets beta — Anthropic platform docs](https://platform.claude.com/docs/en/about-claude/models/whats-new-claude-4-7) [^14]: [Tone and instruction-following changes — Anthropic platform docs](https://platform.claude.com/docs/en/about-claude/models/whats-new-claude-4-7) [^15]: [/ultrareview in Claude Code — Anthropic platform docs](https://platform.claude.com/docs/en/about-claude/models/whats-new-claude-4-7) [^16]: [Claude Opus 4.7 pricing and tokenizer — Anthropic pricing docs](https://platform.claude.com/docs/en/about-claude/pricing) [^17]: [Claude Opus 4.7 safety profile — Anthropic announcement](https://www.anthropic.com/news/claude-opus-4-7) [^18]: [Claude Opus 4.7 System Card — Anthropic](https://www.anthropic.com/news/claude-opus-4-7)