AIHubMixAIHubMix
← Models

claude-opus-5 vs claude-sonnet-5

Output tokens cost $25.00 per million on claude-opus-5 and $10.00 per million on claude-sonnet-5. Input tokens cost $5.00 per million on claude-opus-5 and $2.00 per million on claude-sonnet-5. Cached input tokens are billed at $0.50 per million on claude-opus-5 and $2.00 per million on claude-sonnet-5. Context windows are 1,000,000 tokens on claude-opus-5 and 1,000,000 tokens on claude-sonnet-5. First-token latency measured on AIHubMix is 1.4s on claude-opus-5 and 3.3s on claude-sonnet-5. Measured output throughput is 56.7 tokens/s on claude-opus-5 and 91.0 tokens/s on claude-sonnet-5. On the LMArena coding leaderboard claude-opus-5 scores 1532 and claude-sonnet-5 scores 1526.

Anthropicclaude-opus-5Anthropicclaude-sonnet-5
+ Add model 2/4
Anthropic
claude-opus-5
Anthropic · text, image → text

Claude Opus 5 is Anthropic’s flagship model for demanding reasoning, coding, and long-horizon agentic work. It is particularly strong at end-to-end software tasks, code review and bug finding, visual analysis of charts and documents, complex office deliverables, and coordinating parallel subagents. The model maintains strong instruction following and tool use across extended tasks, while remaining effective at lower effort settings for workloads that prioritize latency and token efficiency.

Input$5.00 /M
Output$25.00 /M
Details
Anthropic
claude-sonnet-5
Anthropic · text, image → text

Claude Sonnet 5 is the next generation of Anthropic's Sonnet model family. It is a drop-in upgrade for Claude Sonnet 4.6 with three behavior changes: adaptive thinking is on by default, manual extended thinking now returns a 400 error (it was deprecated on Claude Sonnet 4.6), and setting sampling parameters (temperature, top_p, top_k) to non-default values returns a 400 error

Input$2.00 /M
Output$10.00 /M
Details

Specs & Pricing

Prices are per million tokens. Latency and throughput are rolling averages measured on AIHubMix.

claude-opus-5
claude-sonnet-5
Input price
$5.00
$2.00
Output price
$25.00
$10.00
Cache read
$0.50
$2.00
Context window
1,000,000
1,000,000
Max output
128,000
128,000
Latency (first token)
1.4 s
3.3 s
Throughput
56.7 TPS
91.0 TPS
Modalities
textimage
textimage
Features
thinkingtoolsstructured outputslong contextweb
thinkingtoolsfunction callingstructured outputs
Endpoints
chat_completions · claude_api · responses
chat_completions · gemini_api · claude_api

Promotional prices show the discounted rate; see each model page for promotion windows.

Usage on AIHubMix last 30 days

Daily traffic served through AIHubMix — how demand for each model is trending.

claude-opus-5claude-sonnet-5

Tokens / day

-

Requests / day

-

Live Performance last 3 days

Measured on real AIHubMix traffic, hourly buckets. Gaps mean no traffic in that hour.

claude-opus-5claude-sonnet-5

Throughput (TPS)

-

Latency (s)

-

Uptime (%)

-

Arena Scores

LMArena ratings by capability (Bradley-Terry, commonly called Elo). Higher is better.

Text
claude-opus-5claude-sonnet-5
1400146015201580
Overall
14621492
Coding
15261532
Math
14711537
Hard prompts
14931520
Instruction following
14631498
Multi-turn
14691486
Creative writing
14291472
Longer query
14821506
Chinese
14971590
English
14741493
Vision
claude-opus-5claude-sonnet-5
126013001340
Overall
12751295
OCR
12931305
Diagram
13201323
Homework
13141341
WebDev Arena
claude-opus-5claude-sonnet-5
1480156016401720
Overall
15421669
React
15481688
HTML
15181668
Gaming
15791750
Simulations
15531725
Data analytics
14961597

Source: LMArena (arena.ai) leaderboard, imported by AIHubMix. Models without published ratings are omitted per chart.

Cost Calculator

Estimate your monthly bill for the same workload on each model.

claude-sonnet-5
$270 /mo
claude-opus-5
$675 /mo

Monthly = daily × 30. Discounted rates applied where a promotion is active.

FAQ

Which is cheaper: claude-opus-5, claude-sonnet-5?

claude-sonnet-5: $10.00/M output tokens; claude-opus-5: $25.00/M. Use the cost calculator above to estimate your own workload.

How do their coding arena scores compare?

claude-opus-5: 1532; claude-sonnet-5: 1526 (LMArena coding leaderboard).

Which responds faster?

claude-opus-5: 1.4s first-token latency measured on AIHubMix; see the live performance charts above for how each model behaves across the day.

How large is each context window?

claude-opus-5 accepts 1,000,000 and claude-sonnet-5 accepts 1,000,000 input tokens. Maximum output per request is 128,000 tokens on claude-opus-5 and 128,000 tokens on claude-sonnet-5.

Which one generates tokens faster?

claude-sonnet-5 at 91.0 tokens/s and claude-opus-5 at 56.7 tokens/s, measured as output throughput on AIHubMix — a separate metric from first-token latency.

What inputs and capabilities does each model support?

claude-opus-5 accepts text and image input and supports thinking, tool calling, structured outputs, long context and web search; claude-sonnet-5 accepts text and image input and supports thinking, tool calling, function calling and structured outputs.

Can I call claude-opus-5 and claude-sonnet-5 with the same API key?

Yes. AIHubMix serves every model on this page behind one OpenAI-compatible endpoint, so switching between them is a one-line change to the model field — no second account, key or SDK.

Popular Comparisons

Related model match-ups readers also look at.

kimi-k3vsqwen3.8-max-previewclaude-opus-5vsgpt-5.6-lunagemini-3.1-pro-previewvsgpt-5.5glm-5.2vskimi-k3gemini-3.6-flashvsqwen3.8-maxdeepseek-v4-flashvsglm-5.2claude-opus-5vsgrok-4.5
AIHubMix

© 2023 - 2026 AIHubMix, LLC

Product

Company

Developer

Connect