Anthropic · Closed-API Model
Claude Opus 4.7: Adaptive Thinking, Tested
Claude Opus 4.7 (April 16, 2026) is Anthropic's flagship reasoning model and the first model with Adaptive Thinking — a mode that auto-tunes thinking compute per request based on task difficulty. Scores 87.6% on SWE-Bench Verified Adaptive harness, 1M context, $5/$25 per million tokens. This review covers when Opus 4.7 is worth the premium over Sonnet 4.6, how Adaptive Thinking actually works, and which open-weight models come closest if you need local deep reasoning.
Update (June 2026): Opus 4.7 is no longer Anthropic's flagship. Claude Opus 4.8 shipped May 28, 2026 and is now the current top model — same $5/$25 per-Mtok pricing and 1M-token context as 4.7, but a higher 88.6% SWE-Bench Verified, plus a new 2.5× “fast mode” and effort control. Opus 4.7 (below) is now the previous generation. Read the full Opus 4.8 review or Anthropic's announcement.
Note: Opus 4.7 is API-only — cannot be downloaded or run locally. For deep reasoning you can self-host, see DeepSeek V4-Pro, Kimi K2.6, and GLM-5.
Key takeaways
- →Adaptive Thinking — model auto-tunes reasoning compute per request, no manual tier selection.
- →87.6% SWE-Bench Verified Adaptive — strong, and ahead of cheaper Sonnet 4.6 (79.6%), though both trail the raw-coding leaders (GPT-5.5 ~88.7%, Opus 4.8 ~88.6%).
- →$5/$25 per Mtok — modestly above Sonnet 4.6 ($3/$15); reserve for hardest 10% of tasks.
- →1M context — same as Sonnet 4.6; works with prompt caching.
- →Best for: research-grade analysis, novel algorithm design, complex multi-step reasoning.
Quick verdict
Use Claude Sonnet 4.6 for ~90% of coding work — it's the best-value frontier-class model: near-flagship quality at roughly 1/8 the price of the top models, fast, with a 1M-token context (79.6% SWE-Bench Verified — slightly below Opus 4.7, well ahead on value). Reach for Opus 4.7 only when the task visibly defeats Sonnet 4.6: complex multi-file refactors, novel algorithm design, deep research analysis, or anywhere Adaptive Thinking earns its keep.
For privacy-required deep-reasoning workloads where you need local hosting, DeepSeek V4-Pro (8× H100) or Kimi K2.6 (also 8× H100) are the realistic alternatives.
Opus 4.8 replaced Opus 4.7 as the flagship
As of June 2026, Claude Opus 4.8 is Anthropic's current flagship, and Opus 4.7 is the previous generation. Anthropic released Opus 4.8 on May 28, 2026, roughly six weeks after 4.7. If you're choosing a model today, default to 4.8 — it's a drop-in upgrade at the same list price.
| Opus 4.8 (current) | Opus 4.7 (previous) | |
|---|---|---|
| Released | May 28, 2026 | April 16, 2026 |
| SWE-Bench Verified | 88.6% | 87.6% |
| SWE-Bench Pro | 69.2% | 64.3% |
| Context window | 1M tokens | 1M tokens |
| API pricing (in/out) | $5 / $25 per Mtok | $5 / $25 per Mtok |
| Notable new features | Fast mode (~2.5×), effort control | — |
Opus 4.8 keeps Adaptive Thinking and the same tools/platform features as 4.7, and adds a research-preview fast mode (set speed: "fast" for up to ~2.5× higher output tokens/sec at premium pricing) plus effort control (high by default, with “extra”/“max” for harder tasks). The rest of this review still describes Opus 4.7 specifically — accurate for the previous generation, useful if you're pinned to it. Sources: Anthropic news · Claude API docs.
Opus 4.7 specs at a glance
| Vendor | Anthropic |
| Release date | April 16, 2026 |
| Model ID | claude-opus-4-7-20260416 |
| Architecture | Dense transformer with Adaptive Thinking |
| Context window | 1,000,000 tokens |
| Max output | 64,000 tokens (excluding thinking trace) |
| Modalities | Text · Code · Vision |
| License | Proprietary |
| Local self-hostable? | No |
Adaptive Thinking explained
Most reasoning models (OpenAI o-series, Gemini 3.1 Pro thinking tiers) make you choose how much thinking to apply per request. Opus 4.7 changes this — the model decides for itself based on task difficulty.
Mechanism: a lightweight pre-pass classifies the input difficulty, then the model spends correspondingly more compute thinking before answering. Easy questions (e.g., “what's 2+2”) get fast responses with no thinking. Medium complexity gets a brief thinking pass. Hard problems trigger extended reasoning over seconds to minutes.
Why it matters: removes the cognitive overhead of picking a thinking tier. For dev tools and agents, Adaptive Thinking means you set the model once and it auto-routes per-task complexity. Cost implication: output tokens include the thinking trace, so a hard problem can multiply effective cost by 2-4× — but you only pay it when needed.
Compare to Gemini 3.1 Pro's explicit Tier 1/2/3 system or GPT-5.5's Instant/Standard/Pro split. Opus 4.7's approach is more elegant but less predictable for cost forecasting.
Benchmarks
| Benchmark | Opus 4.7 | Sonnet 4.6 | GPT-5.5 Pro | Gemini 3.1 Pro |
|---|---|---|---|---|
| SWE-Bench Verified Adaptive | 87.6% | 79.6% | 88.4% | 87.9% |
| GPQA Diamond (PhD science) | 87.3% | 85.7% | 87.6% | 88.2% |
| AIME 2025 (math) | 92.8% | 91.5% | 96.4% | 94.0% |
| ARC-AGI-2 (reasoning) | 71.8% | 68.4% | 73.5% | 77.1% |
| MMLU-Pro | 89.4% | 87.9% | 90.6% | 89.4% |
Pricing & access
API
- Input: $15.00 per 1M tokens
- Output (incl. thinking): $75.00 per 1M tokens
- Cached input: $1.50 per 1M (90% off)
- Batch: 50% off (24h SLA)
Subscription
- Claude Pro: $20/mo — limited Opus 4.7
- Claude Max: $100-200/mo — much higher Opus quota
- Bedrock / Vertex: Same per-token pricing
When to pick Opus 4.7
- ✓Hardest 5-10% of tasks where Sonnet 4.6 visibly struggles.
- ✓Research-grade analysis, novel algorithm design, deep multi-step reasoning.
- ✓Workloads where Adaptive Thinking's auto-routing eliminates manual tier selection.
- ✓Anthropic-stack teams already using Sonnet 4.6; Opus 4.7 is the natural escalation.
FAQ
What is Claude Opus 4.7?
Opus 4.7 vs Sonnet 4.6: when do I use each?
What is Adaptive Thinking?
How much does Claude Opus 4.7 cost?
Can I run Claude Opus 4.7 locally?
Opus 4.7 vs GPT-5.5 Pro: which is better for hard reasoning?
Related models
- → Claude Sonnet 4.6 — your default for ~90% of coding work
- → Claude Fable 5 & Mythos 5 — Anthropic's Mythos tier above Opus, since suspended by export-control order
- → Claude Opus 4.1 — predecessor; lower benchmark scores
- → GPT-5.5 Pro — closest closed alternative; tied on most reasoning
- → Grok 4.3 — xAI's frontier model at far lower $1.25/$2.50 pricing
- → Gemini 3.1 Pro — explicit thinking tiers + 1M context
- → DeepSeek V4-Pro — open-weight deep-reasoning alternative
- → Best AI models May 2026 — pillar comparison
Go from reading about AI to building with AI
20 structured courses. Hands-on projects. Runs on your machine. Start free.
Written by the Local AI Master Team
The team behind Local AI Master
We build Local AI Master around practical, testable local AI workflows: model selection, hardware planning, RAG systems, agents, and MLOps. The goal is to turn scattered tutorials into a structured learning path you can follow on your own hardware.