Back to models

Claude Sonnet 5

FeaturedAvailable Serverless
Anthropic•
Try this modelView documentation

Claude Sonnet 5 brings near-flagship coding and agentic quality to the Sonnet tier, with a 1M-token context window and up to 128K output tokens. It is the default choice for production workloads that need Claude quality without Opus cost. Note that Sonnet 5 uses a new tokenizer that produces roughly 30 percent more tokens for the same text than Sonnet 4.6, so budget accordingly when migrating: the per-token price is the same, but the same document costs more to process. Available on the DOS.AI gateway with one API key and unified billing.

Capabilities:Advanced reasoningCode generationVisionTool callingLong context (1M)
Use cases:Production agentsCode developmentCustomer supportLong-document analysis
Modalities
Input / Output Price
$3.15/$15.75/ 1M
Context Length
1M tokens
Released
Jun 30, 2026

Providers

Different companies host the same model. DOS.AI routes your request to one of them based on the routing mode you pick — Balanced (price + speed), Nitro (fastest), or Exacto (highest tool-calling accuracy).

Reasoning effortⓘAll
ProviderInput /1MOutput /1M
Latency
UptimeContext
Anthropic
$3.15$15.75
1.39 s
100.0%1M

Pricing

Transparent pass-through token pricing per million tokens. DOS.AI strictly enforces a Zero-Markup retail policy.

Input / Prompt
$3.15/ 1M tokens
Published official list price
Output / Completion
$15.75/ 1M tokens
Zero-markup retail rate
Prompt Cache Read
$0.315/ 1M tokens
Up to 90% cache discount
Context Window
1Mtokens
Full context window supported

Observed performance

Provider results over the last 7d.

Updated Sep 19, 05:38 PM UTC
Anthropic
39 samples
Uptime
100.0%
P50 response
1.39 s
P95 response
2.26 s
Latency spread

Uptime is based on observed valid provider attempts. Not an SLA.

Latency is measured to provider response headers, not end-to-end completion.

Uptime & Reliability

Live upstream health and gateway reliability. Automatic multi-provider failover protects your production traffic.

All Endpoints Operational
SLA Target: 99.99%Last 30 Days: 100.0%
30 days ago100.0% uptime over past monthToday

Benchmarks & Evaluations

Standardized evaluations & blind battle ratings. Higher is better — rank percentile reflects position among frontier models on DOS.AI.

Live Gateway Telemetry & Throughput
Window: 7d
Fastest TTFT
-
Time to first token
Median Latency (P50)
1.39 s
Fastest provider response
Tail Latency (P95)
2.26 s
Peak load consistency
Active Endpoints
1 provider
Tested with failover routing
50.4
Top 2%
Artificial Analysis Intelligence Index
Better than 98% of models compared
86.1
Top 1%
Artificial Analysis Coding Index
Better than 99% of models compared
56.7
Top 2%
Artificial Analysis Agentic Index
Better than 98% of models compared

Reasoning

Complex multi-step logic, graduate-level science, and frontier reasoning

3 benchmarks
GPQA Diamondⓘ
91.5%
Graduate-level scientific reasoning
HLEⓘ
51.2%
Humanity's Last Exam frontier benchmark
AA-LCRⓘ
80.5%
Long context reasoning evaluation

Coding

Autonomous repository bug fixing, algorithms, and scientific computation

3 benchmarks
SciCodeⓘ
54.6%
Python programming for scientific computing
SWE-bench Verifiedⓘ
62.5%
Autonomous multi-file bug fixing and test passage
HumanEvalⓘ
91%
Algorithmic code generation pass@1

Knowledge & Reliability

Factual hallucination minimization, tool calling, and domain accuracy

4 benchmarks
AA-Omniscience Accuracyⓘ
60.9%
Proportion of correctly answered questions
AA-Omniscience Non-Hallucination Rateⓘ
37.8%
Rate of avoiding hallucination among non-correct responses
MMLU-Proⓘ
86.1%
Multi-discipline reasoning & higher education knowledge
Exacto BFCL Tool Callingⓘ
99%
Strict schema compliance and function execution

Multimodal & Diagram Reasoning

Multi-discipline diagram perception, document parsing, and visual math

3 benchmarks
MMMUⓘ
71.4%
College-level multi-discipline visual perception
MathVistaⓘ
67.2%
Mathematical problem solving with figures and geometry
DocVQAⓘ
92.4%
Structured tables, charts, and scanned document OCR
Standardized evaluations sourced from Artificial Analysis and open frontier model research benchmarks.

API Integration

Use the DOS API to integrate Claude Sonnet 5 into your applications. Compatible with standard OpenAI SDKs for easy drop-in migration.

</>Run inference
curl https://api.dos.ai/v1/chat/completions \
  -H "Authorization: Bearer $DOS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-5",
    "messages": [
      {
        "role": "user",
        "content": "Hello!"
      }
    ]
  }'
View full documentationGet API key

Explore Related Models

DOS.AI

Auto

DOS.AI Smart Router classifies each request by complexity. Simple and Medium requests use DOS.AI; Complex requests can use an eligible paid route configured in the live catalog.

Usage-based
DOS.AI

DOS

DOS curated meta-model for DOSClaw and first-party DOS products.

$0.07 / 1M tokens
DOS.AI

Qwen3.8-27B

DOS.AI hosted Qwen3.8-27B for reasoning, coding, tool use, and multilingual text generation.

$0.1125 / 1M tokens