Anthropic · Frontier
Claude 4.6 Opus Thinking
Claude Opus 4.6 with adaptive thinking enabled (default effort: high).
- Reasoning
- Tools
- Vision
- Structured output
- PDF input
- Input
- Text, Images
- Output
- Text
- Reasoning effort
- low · medium · high · xhigh · max
- Discovery tier
- Frontier
Pricing
USD per 1M tokens- Input
- $5.00
- Output
- $25.00
- Cache read
- $0.50
Reference prices as published by the model provider. API example
Specifications
- Context window
- 1M
- Max output
- 128K
- Average output (7 days)
- 1.4K tokens
- Longer than 83% of 465 models
Side by side
Claude 4.6 Opus Thinking compared
Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.
| Claude 4.6 Opus ThinkingAnthropicThis model | Claude Opus 4.8 ThinkingAnthropicRemove | |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Discovery tier | Frontier | Frontier |
| Added | Feb 5, 2026 | May 28, 2026 |
| Limits | ||
| Context window | 1M | 1M |
| Max output | 128K | 128K |
| Average output | 1.4K tokens | 1.3K tokens |
| Pricing, USD per 1M tokens | ||
| Input | $5.00 | $5.00 |
| Output | $25.00 | $25.00 |
| Cache read | $0.50 | $0.50 |
| Performance | ||
| Throughput | 77.0 tok/s | 160.0 tok/sBest |
| Time to first token | 1.9 sBest | 3.4 s |
| Request success (24h) | 100.0% | 100.0% |
| Benchmark indexes | ||
| Intelligence Index | 31.9 | 41.8Best |
| Coding Index | — | 74.3 |
| Capabilities | ||
| Reasoning | Yes | Yes |
| Tools | Yes | Yes |
| Vision | Yes | Yes |
| Structured output | Yes | Yes |
| PDF input | Yes | Yes |
| Reasoning effort | low, medium, high, xhigh, max | low, medium, high, xhigh, max |
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
- Claude 4.6 Opus Thinking
- Claude Opus 4.8 Thinking
Intelligence Index
- Claude 4.6 Opus Thinking31.9
- Claude Opus 4.8 Thinking41.8
Claude 4.6 Opus Thinking tested at max reasoning effort
Claude 4.6 Opus Thinking: Better than 74% of 326 scored models
Indexes
Coding Index
Composite of the coding evaluations
- Claude 4.6 Opus Thinking: —
- Claude Opus 4.8 Thinking: 74.3
Agentic work
AutomationBench-AA
Workflow automation, penalized for guardrail violations
- Claude 4.6 Opus Thinking: —
- Claude Opus 4.8 Thinking: 45.6%
AutomationBench-AA tasks completed
Workflows finished end to end without a guardrail violation
- Claude 4.6 Opus Thinking: —
- Claude Opus 4.8 Thinking: 17.4%
AA-Briefcasev1.1
Agentic knowledge work, rated head to head
- Claude 4.6 Opus Thinking: —
- Claude Opus 4.8 Thinking: 1,319 Elo
GDPval-AAv2.1
Economically valuable occupational tasks, rated head to head
- Claude 4.6 Opus Thinking: —
- Claude Opus 4.8 Thinking: 1,454 Elo
Document reasoning
GDP.pdf
Professional reasoning over PDF documents, all-pass rate
- Claude 4.6 Opus Thinking: —
- Claude Opus 4.8 Thinking: 22.8%
AA-LCRv1.1
Reasoning across long documents
- Claude 4.6 Opus Thinking: 78.0%Top
- Claude Opus 4.8 Thinking: 77.7%
MLCR-AA
Long-context reasoning over medical records
- Claude 4.6 Opus Thinking: —
- Claude Opus 4.8 Thinking: 19.4%
Reasoning and knowledge
Humanity’s Last Exam
Expert-level questions across academic fields
- Claude 4.6 Opus Thinking: 39.9%
- Claude Opus 4.8 Thinking: 48.7%Top
GPQA Diamond
Graduate-level science questions
- Claude 4.6 Opus Thinking: 89.6%
- Claude Opus 4.8 Thinking: 92.0%Top
IFBench
Following precise output instructions
- Claude 4.6 Opus Thinking: 53.1%
- Claude Opus 4.8 Thinking: 62.2%Top
Coding
Terminal-Benchv4.0
Practical software engineering in a terminal
- Claude 4.6 Opus Thinking: —
- Claude Opus 4.8 Thinking: 21.7%
SciCode
Scientific computing in Python
- Claude 4.6 Opus Thinking: —
- Claude Opus 4.8 Thinking: 54.4%
Legacy evaluations
Terminal-Bench Hard
Earlier terminal task suite
- Claude 4.6 Opus Thinking: 46.2%
- Claude Opus 4.8 Thinking: 58.3%Top
τ²-Bench Telecom
Tool use in a customer-support simulation
- Claude 4.6 Opus Thinking: 92.1%
- Claude Opus 4.8 Thinking: 94.4%Top
AA-LCR (unversioned)
Long-context reasoning, earlier grading
- Claude 4.6 Opus Thinking: 78.0%Top
- Claude Opus 4.8 Thinking: 77.7%
Scores from Artificial Analysis, updated Oct 26, 2026. Claude 4.6 Opus Thinking was tested at max reasoning effort; compare scores measured at the same effort.Artificial Analysis
- Claude 4.6 Opus Thinking
- Claude Opus 4.8 Thinking
| Leaderboard | Claude 4.6 Opus Thinking | Claude Opus 4.8 Thinking |
|---|---|---|
| ASCII Art | #15 · 1252 | Not in top 20 |
Ranks among Noumen catalog models on Design Arena leaderboards. Only top-20 placements are listed.Design Arena
Live routing data
Performance and reliability
Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.
- Throughput
- 77.0 tok/s
- Faster than 50% of 250 models
- Time to first token
- 1.9 s
- Quicker than 79% of 250 models
- Prompt cache hits
- 45.7%
- About $2.95 per 1M input tokens with caching
- Request success (24h)
- 100.0%
- Reliable
Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.
- Healthy (98% or more)
- Degraded (90–98%)
- Down (under 90%)
- No requests
Developers
Use it through the API
Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.
curl https://www.noumen.xyz/api/v1/chat/completions \
-H "Authorization: Bearer $NOUMEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "anthropic/claude-opus-4.6:thinking",
"reasoning_effort": "medium",
"messages": [{ "role": "user", "content": "Explain this stack trace." }]
}'import os
from openai import OpenAI
client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
model="anthropic/claude-opus-4.6:thinking",
messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)Anthropic
Related models
Other Anthropic models, starting with the same family.