Claude Opus 5.5 vs DeepSeek V4 Pro 0813

At a glance

• Capability index

  • Claude Opus 5.5167.3
  • DeepSeek V4 Pro 0813155.3

  Claude Opus 5.5 is 12.0 points higher.

• Price per 1M tokens

  • Claude Opus 5.5$8.00
  • DeepSeek V4 Pro 0813$1.98

  DeepSeek V4 Pro 0813 is 4 times cheaper.

• Context window

  • Claude Opus 5.51M tokens
  • DeepSeek V4 Pro 08131M tokens

  They take in about the same.

• Released

  • Claude Opus 5.5Sep 22, 2026
  • DeepSeek V4 Pro 0813Aug 12, 2026

  Claude Opus 5.5 is 41 days newer.

Where each scores higher

On the 13 tests both have taken, each on its own scale. A gap of a point or two can sit within a test’s margin of error.

9 more tests have results for only one of them; the table below lists every result.

Everything side by side

Comparison of Claude Opus 5.5 vs DeepSeek V4 Pro 0813

Measure	Claude Opus 5.5Anthropic	DeepSeek V4 Pro 0813DeepSeek
Model
Maker	Anthropic	DeepSeek
Sold here by	Anthropic	CrofAI
API model ID	claude-opus-5-5	deepseek-v4-pro-0813
Released	Sep 22, 2026	Aug 12, 2026
Knowledge cutoff	Jun 2026	Not reported
Weights	Closed: hosted access only	Open: downloadable
Developer’s country	United States	China
Hosts selling it	35including Anthropic directly	42
Training compute	Not published	9.7 × 10²⁴ FLOPEpoch AI estimate
Test results
Capability index	167.3 (best of these models)#1 of 274	155.3#29 of 274
LiveBench overall	83.2% (best of these models)#2 of 66	77.4%#23 of 66
Arena text	1504#4 of 413	Not tested
LiveBench coding	89.3% (best of these models)#2 of 66	77.2%#38 of 66
Arena WebDev	1815#1 of 138	Not tested
Arena coding	1539#12 of 408	Not tested
LiveBench mathematics	97.1% (best of these models)#1 of 66	95.1%#16 of 66
Mock AIME 2024–2025	100.0% (best of these models)#1 of 297	98.6%#20 of 297
FrontierMath Tiers 1–3	91.2% (best of these models)#3 of 114	64.6%#32 of 114
FrontierMath Tier 4	95.0% (best of these models)#6 of 70	26.8%#39 of 70
Arena math	1511#8 of 396	Not tested
SimpleQA Verified	72.2% (best of these models)#4 of 86	52.9%#21 of 86
LiveBench agentic coding	71.7% (best of these models)#2 of 66	55.0%#28 of 66
CursorBench	57.8%#1 of 62	Not tested
LiveBench language	86.3% (best of these models)#10 of 66	82.1%#26 of 66
LiveBench instructions	67.0%#39 of 66	67.7% (best of these models)#36 of 66
Arena creative writing	1516#2 of 411	Not tested
Arena instruction following	1517#2 of 413	Not tested
LiveBench data analysis	80.3% (best of these models)#7 of 66	79.2%#18 of 66
LiveBench reasoning	92.2% (best of these models)#3 of 66	85.8%#34 of 66
GPQA Diamond	90.6%#40 of 319	91.7% (best of these models)#29 of 319
Arena hard prompts	1534#2 of 413	Not tested
Chess puzzles	Not tested	47.0%#14 of 227
Price per million tokens
Input	$4.00	$0.35 (best of these models)
Output	$20.00	$0.80 (best of these models)
Blended, 3 input to 1 output	$8.00	$0.462 (best of these models)
Cached input	$0.20	$0.01 (best of these models)
Writing to the cache	$5.00	Not reported
The maker’s own price	This listing	Not sold directly
Limits
Context window	1M tokens	1M tokens
Max input	1M tokens	1M tokens
Max output	128K tokens	131.1K tokens (best of these models)
Features
Reads	images, PDFs and text	text
Produces	text	text
Reasoning	Yeseffort low, medium, high, xhigh, max; can be switched off	Yeseffort none, low, high, max; can be switched off; thinking budget can be set
Tool calling	Yes	Yes
Structured output	Yes, such as JSON	Yes, such as JSON
File attachments	Yes	No
Temperature setting	Fixed by the provider	Supported

Bold marks the better value in each row. Prices are each model’s own list price, or the middle price across hosts when the maker does not sell it, blended as three input tokens for every output token. Test results as published by LiveBench (Apache-2.0), Epoch AI (CC BY 4.0), Cursor CursorBench, compiled by Epoch AI (CC BY 4.0), Arena (CC BY 4.0); model details and prices from models.dev (MIT). Confirm pricing with the provider before use.