ZMIME
Comparison · 3 models · Updated Nov 18, 2026

o3-pro vs DeepSeek R1 0528 vs Claude Opus 4.1

DeepSeek R1 0528 comes out ahead, 55 to 52 and 51 on our weighted score, and it is the cheaper option too.

Price in rating: Included · Exclude price

  1. OpenAI

    o3-pro

    Released Jun 10, 2025

    52/100
    • ECI147.5
    • Price$20.00 / $80.00
    • Context200K
  2. Our pick

    DeepSeek

    DeepSeek R1 0528

    Released May 28, 2025

    55/100
    • ECI140.8
    • Price$0.512 / $2.19
    • Context164K
  3. Anthropic

    Claude Opus 4.1

    Released Aug 5, 2025

    51/100
    • ECI144.8
    • Price$15.00 / $75.00
    • Context200K
01 — Verdict

DeepSeek R1 0528 is our pick

DeepSeek R1 0528 is the better all-round choice, scoring 55/100 against o3-pro (52) and Claude Opus 4.1 (51). It leads on price. o3-pro wins on capability. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

  • Capabilityo3-proCapabilities Index (ECI): o3-pro 147.5 · Claude Opus 4.1 144.8 · DeepSeek R1 0528 140.8
  • Lowest priceDeepSeek R1 0528DeepSeek R1 0528 $0.932 · Claude Opus 4.1 $30.00 · o3-pro $35.00 per 1M tokens (3:1 blend)
  • Longest contexto3-pro and Claude Opus 4.1o3-pro 200,000 · Claude Opus 4.1 200,000 · DeepSeek R1 0528 163,840 tokens
  • Widest inputsClaude Opus 4.1o3-pro: Text, Images · DeepSeek R1 0528: Text · Claude Opus 4.1: Text, Images, PDFs
  • Self-hostingDeepSeek R1 0528Publishes downloadable weights (MIT)
How the score is built
MeasureWeighto3-proDeepSeek R1 0528Claude Opus 4.1
CapabilityCapabilities Index (ECI)50%756972
Price25%0490
Inputs & features15%703570
Context window10%393639
Overall100%52/10055/10051/100
02 — Side by side

Every spec in one table

Highlighted cells lead their row. Dashes mean the data is not published.

o3-pro vs DeepSeek R1 0528 vs Claude Opus 4.1 specifications side by side
Specificationo3-proOpenAIDeepSeek R1 0528DeepSeekClaude Opus 4.1Anthropic
Capability
Capabilities Index (ECI)147.5 (best)140.8144.8
ECI rank#79 of 158 (best)#113 of 158#93 of 158
GPQA DiamondGraduate-level science questions—76.3%77.3% (best)
FrontierMath Tiers 1–3Research-level mathematics——12.6%
OTIS Mock AIME 2024–2025Competition mathematics—66.4%68.9% (best)
SWE-bench VerifiedFixing real GitHub issues——73.4%
Price per million tokens
Input$20.00$0.512 (best)$15.00
Output$80.00$2.19 (best)$75.00
Cached input———
Blended (3:1)$35.00$0.932 (best)$30.00
Long-context rateSame rateSame rateSame rate
Price sourceOfficial OpenAI APIMedian of 16 providersMedian of 18 providers
Limits
Context window200,000 tokens (best)163,840 tokens200,000 tokens (best)
Max output100,000 tokens (best)32,768 tokens32,000 tokens
Inputs and features
TextYesYesYes
ImagesYesNoYes
PDFsNoNoYes
AudioNoNoNo
VideoNoNoNo
ReasoningYeslow · medium · highYesYes
Tool callingYesYesYes
Structured outputYesNoNo
Availability
WeightsProprietaryOpenMITProprietary
API model IDo3-pro——
API providers101619 (best)
ReleasedJun 10, 2025May 28, 2025Aug 5, 2025
Knowledge cutoffMay 2024—Mar 31, 2025
03 — Cost

What would a month cost?

Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.

  • o3-pro$360.00
  • DeepSeek R1 0528$9.50
  • Claude Opus 4.1$300.00
04 — Questions

Which should you choose?

Which is better: o3-pro, DeepSeek R1 0528 or Claude Opus 4.1?

DeepSeek R1 0528 is the better all-round choice, scoring 55/100 against o3-pro (52) and Claude Opus 4.1 (51). It leads on price. o3-pro wins on capability. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

Which is cheaper, o3-pro, DeepSeek R1 0528 or Claude Opus 4.1?

DeepSeek R1 0528 is cheaper at $0.512 input / $2.19 output per million tokens (median across 16 API providers). Claude Opus 4.1 costs $15.00 input / $75.00 output per million tokens (median across 18 API providers); o3-pro costs $20.00 input / $80.00 output per million tokens (official OpenAI API price). At a typical mix of three input tokens to one output token, that is $0.932 per million tokens for DeepSeek R1 0528 versus $30.00 for Claude Opus 4.1 (32× as much) and $35.00 for o3-pro (38× as much).

Which scores higher on benchmarks?

o3-pro scores higher on the Capabilities Index (ECI): o3-pro 147.5 (#79 of 158), Claude Opus 4.1 144.8 (#93 of 158) and DeepSeek R1 0528 140.8 (#113 of 158). The confidence ranges of the top two overlap (145.6–149.4 vs 143.0–146.5), so treat the gap as small.

Which is better for coding?

There are no published SWE-bench Verified results for o3-pro and DeepSeek R1 0528 yet, so there is no like-for-like coding score. On overall capability, o3-pro leads, which tends to carry over to coding, but test on your own codebase. All three support tool calling for agent workflows.

Which has the bigger context window?

o3-pro and Claude Opus 4.1 have the largest context windows (200,000 and 200,000 tokens), against 163,840 for DeepSeek R1 0528. Maximum output per response: o3-pro up to 100,000, DeepSeek R1 0528 up to 32,768, Claude Opus 4.1 up to 32,000 tokens.

Which can read images, PDFs, audio or video?

o3-pro accepts text and images; DeepSeek R1 0528 accepts text; Claude Opus 4.1 accepts text, images and PDFs. Claude Opus 4.1 handles the widest range of inputs.

Are any of these open source?

DeepSeek R1 0528 publishes its weights (MIT) and can be self-hosted; o3-pro and Claude Opus 4.1 is proprietary.

Which is newer?

Claude Opus 4.1 is the newest, released Aug 5, 2025. o3-pro came out Jun 10, 2025; DeepSeek R1 0528 came out May 28, 2025. Knowledge cutoff: o3-pro May 2024, Claude Opus 4.1 Mar 31, 2025.

How do you decide the winner?

This rating uses capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Nov 18, 2026.