NOUMENOpen app
NOUMEN

Anthropic · Frontier

Claude Sonnet 5

Added Jun 30, 2026

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Use in the app

Claude Sonnet 5 supports coding, agentic workflows, long-context reasoning, computer use, and practical knowledge work.

  • Tools
  • Vision
  • Structured output
  • PDF input
Input
Text, Images
Output
Text
Discovery tier
Frontier

Pricing

USD per 1M tokens
Input
$2.00
Output
$10.00
Cache read
$0.20
Cache write
$2.50
$4.00 for a 1-hour cache

Reference prices as published by the model provider. API example

Specifications

Context window
1M
Max output
128K
Average output (7 days)
291 tokens
Longer than 30% of 473 models

Side by side

Claude Sonnet 5 compared

Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Claude Sonnet 5AnthropicThis modelGrok 4.5xAIRemove
ProviderAnthropicxAI
Discovery tierFrontierFrontier
AddedJun 30, 2026Jul 8, 2026
Limits
Context window1MBest500K
Max output128K—
Average output291 tokens1K tokens
Pricing, USD per 1M tokens
Input$2.00$2.00
Output$10.00$6.00Best
Cache read$0.20Best$0.30
Cache write$2.50—
Performance
Throughput102.8 tok/sBest70.5 tok/s
Time to first token12.1 s1.7 sBest
Request success (24h)99.8%100.0%Best
Benchmark indexes
Intelligence Index38.238.8Best
Coding Index71.572.4Best
Capabilities
ReasoningNoYes
ToolsYesYes
VisionYesYes
Structured outputYesYes
PDF inputYesNo
Reasoning effort—low, medium, high

Public evaluations

Benchmarks

Independent evaluation results, with how each score ranks among the models in the Noumen catalog.

  • Claude Sonnet 5
  • Grok 4.5

Intelligence Index

  • Claude Sonnet 538.2
  • Grok 4.538.8

Claude Sonnet 5 tested at max reasoning effort

Claude Sonnet 5: Better than 82% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • Claude Sonnet 5: 71.5
    • Grok 4.5: 72.4Top

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Claude Sonnet 5: 36.5%
    • Grok 4.5: 57.9%Top

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Claude Sonnet 5: 10.8%
    • Grok 4.5: 26.3%Top

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Claude Sonnet 5: 1,357 EloTop
    • Grok 4.5: 1,278 Elo

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Claude Sonnet 5: 1,464 EloTop
    • Grok 4.5: 1,381 Elo

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Claude Sonnet 5: 13.2%
    • Grok 4.5: 18.8%Top

  • AA-LCRv1.1

    Reasoning across long documents

    • Claude Sonnet 5: 82.0%Top
    • Grok 4.5: 79.3%

  • MLCR-AA

    Long-context reasoning over medical records

    • Claude Sonnet 5: 5.0%
    • Grok 4.5: 16.1%Top

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Claude Sonnet 5: 41.3%
    • Grok 4.5: 42.7%Top

  • GPQA Diamond

    Graduate-level science questions

    • Claude Sonnet 5: 91.1%
    • Grok 4.5: 93.1%Top

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Claude Sonnet 5: 14.1%Top
    • Grok 4.5: 10.6%

  • SciCode

    Scientific computing in Python

    • Claude Sonnet 5: 54.3%
    • Grok 4.5: 55.0%Top

Legacy evaluations

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Claude Sonnet 5: 82.0%Top
    • Grok 4.5: 79.3%

Scores from Artificial Analysis, updated Oct 26, 2026. Claude Sonnet 5 was tested at max reasoning effort; compare scores measured at the same effort.Artificial Analysis

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
102.8 tok/s
Faster than 63% of 256 models
Time to first token
12.1 s
Quicker than 6% of 256 models
Prompt cache hits
73.9%
About $0.67 per 1M input tokens with caching
Request success (24h)
99.9%
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.

Hourly request success · last 72 hoursLowest hour 99.0%
  • 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.

Read the quickstart
curl https://www.noumen.xyz/api/v1/chat/completions \
  -H "Authorization: Bearer $NOUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-sonnet-5",
    "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-sonnet-5",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Anthropic

Other Anthropic models, starting with the same family.