ZMIME
Comparison · 3 models · Updated Nov 18, 2026

GPT-5 vs DeepSeek-R1 vs Gemini 2.5 Pro

Gemini 2.5 Pro comes out ahead, 64 to 60 and 50 on our weighted score, though DeepSeek-R1 is 40% cheaper per token.

Price in rating: Included · Exclude price

  1. OpenAI

    GPT-5

    Released Aug 7, 2025

    60/100
    • ECI150.0
    • Price$1.25 / $10.00
    • Context400K
  2. DeepSeek

    DeepSeek-R1

    Released Jan 20, 2025

    50/100
    • ECI138.6
    • Price$1.02 / $5.14
    • Context128K
  3. Our pick

    Google

    Gemini 2.5 Pro

    Released Jun 17, 2025

    64/100
    • ECI144.9
    • Price$1.25 / $10.00
    • Context1.05M
01 — Verdict

Gemini 2.5 Pro is our pick

Gemini 2.5 Pro is the better all-round choice, scoring 64/100 against GPT-5 (60) and DeepSeek-R1 (50). It leads on inputs & features and context window. GPT-5 wins on capability. DeepSeek-R1 wins on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

  • CapabilityGPT-5Capabilities Index (ECI): GPT-5 150.0 · Gemini 2.5 Pro 144.9 · DeepSeek-R1 138.6
  • Lowest priceDeepSeek-R1DeepSeek-R1 $2.05 · GPT-5 $3.44 · Gemini 2.5 Pro $3.44 per 1M tokens (3:1 blend)
  • Longest contextGemini 2.5 ProGemini 2.5 Pro 1,048,576 · GPT-5 400,000 · DeepSeek-R1 128,000 tokens
  • Widest inputsGemini 2.5 ProGPT-5: Text, Images · DeepSeek-R1: Text · Gemini 2.5 Pro: Text, Images, PDFs, Audio, Video
  • Self-hostingDeepSeek-R1Publishes downloadable weights
How the score is built
MeasureWeightGPT-5DeepSeek-R1Gemini 2.5 Pro
CapabilityCapabilities Index (ECI)50%776773
Price25%233423
Inputs & features15%7035100
Context window10%503265
Overall100%60/10050/10064/100
02 — Side by side

Every spec in one table

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

GPT-5 vs DeepSeek-R1 vs Gemini 2.5 Pro specifications side by side
SpecificationGPT-5OpenAIDeepSeek-R1DeepSeekGemini 2.5 ProGoogle
Capability
Capabilities Index (ECI)150.0 (best)138.6144.9
ECI rank#68 of 158 (best)#118 of 158#92 of 158
GPQA DiamondGraduate-level science questions86.2% (best)71.7%85.3%
FrontierMath Tiers 1–3Research-level mathematics55.4% (best)—24.6%
OTIS Mock AIME 2024–2025Competition mathematics91.4% (best)53.3%84.7%
SWE-bench VerifiedFixing real GitHub issues73.6% (best)—57.6%
SimpleQA VerifiedShort factual questions50.1%——
Price per million tokens
Input$1.25$1.02 (best)$1.25
Output$10.00$5.14 (best)$10.00
Cached input$0.125—$0.125
Blended (3:1)$3.44$2.05 (best)$3.44
Long-context rateSame rateSame rateOver 200K: $2.50 / $15.00
Price sourceOfficial OpenAI APIMedian of 12 providersOfficial Google API
Limits
Context window400,000 tokens128,000 tokens1,048,576 tokens (best)
Max output128,000 tokens (best)32,768 tokens65,536 tokens
Inputs and features
TextYesYesYes
ImagesYesNoYes
PDFsNoNoYes
AudioNoNoYes
VideoNoNoYes
ReasoningYesminimal · low · medium · highYesYes
Tool callingYesYesYes
Structured outputYesNoYes
Availability
WeightsProprietaryOpenProprietary
API model IDgpt-5—gemini-2.5-pro
API providers35 (best)1331
ReleasedAug 7, 2025Jan 20, 2025Jun 17, 2025
Knowledge cutoffSep 30, 2024Jul 2024Jan 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.

  • GPT-5$32.50
  • DeepSeek-R1$20.53
  • Gemini 2.5 Pro$32.50
04 — Questions

Which should you choose?

Which is better: GPT-5, DeepSeek-R1 or Gemini 2.5 Pro?

Gemini 2.5 Pro is the better all-round choice, scoring 64/100 against GPT-5 (60) and DeepSeek-R1 (50). It leads on inputs & features and context window. GPT-5 wins on capability. DeepSeek-R1 wins on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

Which is cheaper, GPT-5, DeepSeek-R1 or Gemini 2.5 Pro?

DeepSeek-R1 is cheaper at $1.02 input / $5.14 output per million tokens (median across 12 API providers). GPT-5 costs $1.25 input / $10.00 output per million tokens (official OpenAI API price); Gemini 2.5 Pro costs $1.25 input / $10.00 output per million tokens (official Google API price). At a typical mix of three input tokens to one output token, that is $2.05 per million tokens for DeepSeek-R1 versus $3.44 for GPT-5 (1.7× as much) and $3.44 for Gemini 2.5 Pro (1.7× as much).

Which scores higher on benchmarks?

GPT-5 scores higher on the Capabilities Index (ECI): GPT-5 150.0 (#68 of 158), Gemini 2.5 Pro 144.9 (#92 of 158) and DeepSeek-R1 138.6 (#118 of 158). On individual benchmarks: GPQA Diamond — GPT-5 86.2%, Gemini 2.5 Pro 85.3%, DeepSeek-R1 71.7%; OTIS Mock AIME 2024–2025 — GPT-5 91.4%, Gemini 2.5 Pro 84.7%, DeepSeek-R1 53.3%.

Which is better for coding?

There are no published SWE-bench Verified results for DeepSeek-R1 yet, so there is no like-for-like coding score. On overall capability, GPT-5 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?

Gemini 2.5 Pro has the largest context window at 1,048,576 tokens, against 400,000 for GPT-5 and 128,000 for DeepSeek-R1. Maximum output per response: GPT-5 up to 128,000, DeepSeek-R1 up to 32,768, Gemini 2.5 Pro up to 65,536 tokens.

Which can read images, PDFs, audio or video?

GPT-5 accepts text and images; DeepSeek-R1 accepts text; Gemini 2.5 Pro accepts text, images, PDFs, audio and video. Gemini 2.5 Pro handles the widest range of inputs.

Are any of these open source?

DeepSeek-R1 publishes its weights and can be self-hosted; GPT-5 and Gemini 2.5 Pro is proprietary.

Which is newer?

GPT-5 is the newest, released Aug 7, 2025. Gemini 2.5 Pro came out Jun 17, 2025; DeepSeek-R1 came out Jan 20, 2025. Knowledge cutoff: GPT-5 Sep 30, 2024, DeepSeek-R1 Jul 2024, Gemini 2.5 Pro Jan 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.