Skip to main content
BenchLM
Data

DeepSeek V4 Pro vs GPT-5

Updated November 18, 2026. Rank says DeepSeek V4 Pro is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.

Share or export
Share on XLinkedInAPI/MCP

Decision reading

DeepSeek V4 Pro has the higher public point estimate, 62.71 versus 48.96. Their conditional score ranges overlap. These ranges do not establish rank confidence. 5 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.

Model A
DeepSeek logo

DeepSeek

62.71/100

Estimated · Public rank #39

Conditional range 48.4–77.1

Model B
OpenAI logo

OpenAI

48.96/100

Estimated · Public rank #109

Conditional range 34.6–63.3

API price per 1M tokens

DeepSeek V4 Pro: $1.32 input / $3.96 output

GPT-5: $1.25 input / $10 output

Context window

DeepSeek V4 Pro: 1M

GPT-5: 400K

Released

DeepSeek V4 Pro: August 13, 2026

GPT-5: August 7, 2025

Shared results
5
DeepSeek V4 Pro only
40
GPT-5 only
1
Like-for-like categories
1 / 8
Estimated: DeepSeek V4 Pro and GPT-5. Conditional ranges do not establish rank confidence.How the comparison works

Which one for your work

Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.

  • Long documents

    Prompts that approach the documented context limit

    DeepSeek V4 Pro

    DeepSeek V4 Pro has the larger documented context window.

    Confidence: documented
  • Chat turn cost

    1K fresh input + 500 output tokens

    DeepSeek V4 Pro

    DeepSeek V4 Pro has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates
  • Cache-heavy agent loop cost

    200K cached + 20K fresh input + 10K output tokens

    DeepSeek V4 Pro

    DeepSeek V4 Pro has the lower estimated token cost for this stated workload. GPT-5 has no published cached-input rate, so cached tokens use its listed input rate.

    Confidence: rate-fallback
Show secondary and unsupported calls
  • Repository review cost

    50K fresh input + 3K output tokens

    DeepSeek V4 Pro

    DeepSeek V4 Pro has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates
  • Coding work

    Code generation, repair, and software-engineering tasks

    Not enough matched evidence

    GPT-5 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.

    Confidence: limited
  • Agentic work

    Tool use, computer use, and multi-step task completion

    Not enough matched evidence

    GPT-5 is not ranked on the public lane for agentic, so no winner is named for agentic.

    Confidence: limited

Which one for a specific job

Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.

The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.

51.4DeepSeek V4 Pro36.7GPT-5

Directional only · BenchAlign v5.8

DeepSeek V4 Pro has the higher coding point estimate. Conditional score ranges do not establish rank confidence.

Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.

Same basis rules as the category table below

Or describe your own job

The LLM Selector reads what you need a model for and returns a shortlist, ranked in the same order the job pages publish.

Describe the task and constraints. This selector recommends models; it cannot carry out the task or access your files and accounts.

We keep what you type here, drafts included, for 180 days, and after that only the job and requirements read from it. Nothing is tied to your IP address or account. Leave out names and other personal details. Privacy

What is actually comparable

Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.

3 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.

Shape of the matched evidence

Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.

Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.

Bars run 0–100 on each benchmark’s normalized display scale

Category results, on a stated basis

Each row shows the public-lane category score for both models: the BenchAlign v5.8 lane for agentic, coding, reasoning, multimodal, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.

Knowledge

Like-for-like
DeepSeek V4 Pro
64.2
Supported · #35/179
GPT-5
53.1
Supported · #67/179
Basis
BenchAlign v5.8 lane · 7 vs 2 public rows
Reading
DeepSeek V4 Pro leads · intervals overlap

Coding

Directional only
DeepSeek V4 Pro
51.4
Supported · #41/160
GPT-5
36.7
Estimated · #85/160
Basis
BenchAlign v5.8 lane · 15 vs 3 public rows
Reading
Directional only

Reasoning

Directional only
DeepSeek V4 Pro
58.1
Supported · #33/159
GPT-5
43.7
Estimated · #75/159
Basis
BenchAlign v5.8 lane · 4 vs 0 public rows
Reading
Directional only

Instruction following

Directional only
DeepSeek V4 Pro
93.8
#7/157
GPT-5
89.4
#35/157
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Directional only

Agentic

Not comparable
DeepSeek V4 Pro
57.0
Supported · #41/138
GPT-5
Not ranked
Basis
BenchAlign v5.8 lane · 13 vs 1 public rows
Reading
Not comparable

Multimodal

Not comparable
DeepSeek V4 Pro
Not ranked
GPT-5
Not ranked
Basis
BenchAlign v5.8 lane · 0 vs 0 public rows
Reading
Not comparable

Multilingual

Not comparable
DeepSeek V4 Pro
Not ranked
GPT-5
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Math

Not comparable
DeepSeek V4 Pro
73.9
#4/13
GPT-5
Not ranked
Basis
Provisional lane · 3 vs 0 weighted rows
Reading
Not comparable

Ranks count the models scored in each category’s lane, so the agentic, coding, reasoning, multimodal, and knowledge denominators (BenchAlign v5.8) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.

Supported evidence per lane · bars run 0–100Methodology

What each workload costs

Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.

Chat turn

1K fresh input + 500 output tokens

DeepSeek V4 Pro
$0.0033
Fits in one request
GPT-5
$0.00625
Fits in one request

DeepSeek V4 Pro has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

DeepSeek V4 Pro
$0.07788
Fits in one request
GPT-5
$0.0925
Fits in one request

DeepSeek V4 Pro has the lower modeled cost

Costs use the listed standard API rates.

Cache-heavy agent loop

200K cached + 20K fresh input + 10K output tokens

DeepSeek V4 Pro
$0.0748
Fits in one request
GPT-5
$0.375
Fits in one request
Cached input priced at the published list-input rate

DeepSeek V4 Pro has the lower modeled cost

GPT-5 has no published cached-input rate, so cached tokens use its listed input rate.

Your workload

Enter a month of tokens to price it at both models’ listed API rates.

DeepSeek V4 Pro
$21.12
per month
GPT-5
$32.50
per month

Standard listed rates, no batch or volume discounts.

Cached input falls back to the list input rate only where a cached rate is unpublished

Price against score, among all priced models

Overall score against output price per 1M tokens, one model per family. Pricing as of November 18, 2026.

DeepSeek V4 ProGPT-5

20406080100$0.10$0.50$1$5$10$50$100Output price, $ per 1M tokens (log)Overall score
  • 4 cheaper models score at least as high as DeepSeek V4 Pro. The nearest is Gemini 3.8 Flash (score 64.4, $3.75 per 1M output tokens).
  • 31 models score at least as high as GPT-5 for the same price or less. The nearest is Gemini 4 Carbon (score 85.0, also $10.00 per 1M output tokens).

Gray dots are the other priced models. The dark step line is the efficiency frontier: for a model under it, another model costs no more and scores at least as high.

See DeepSeek V4 Pro on the full price vs performance chart

Specification differences

Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.

Cached-input rate

A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.

DeepSeek V4 Pro

$0.044 per 1M cached input tokens

DeepSeek: Models & Pricing

GPT-5

Not published

Reasoning profile

DeepSeek V4 Pro

Reasoning

GPT-5

Reasoning

Weight access

DeepSeek V4 Pro

Open Weight

GPT-5

Proprietary

License

DeepSeek V4 Pro

Open Weight

GPT-5

Proprietary

Release date

DeepSeek V4 Pro

2026-08-13

GPT-5

2025-08-07

If you already use one of these models

Deployment change
The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
Quality signal
DeepSeek V4 Pro has the higher public point estimate, 62.71 versus 48.96. Their conditional score ranges overlap. These ranges do not establish rank confidence.
Workload cost
Repository review: $0.07788 vs $0.0925. Cache-heavy agent loop: $0.0748 vs $0.375.
Context tradeoff
DeepSeek V4 Pro has the larger documented window (1M).
Run the same representative tasks against both endpoints before changing production traffic.

Questions

Which is better, DeepSeek V4 Pro or GPT-5?

DeepSeek V4 Pro has the higher public point estimate, 62.71 versus 48.96. Their conditional score ranges overlap. These ranges do not establish rank confidence. The page therefore keeps the decision tied to the specific documented workload.

Which is better for coding, DeepSeek V4 Pro or GPT-5?

DeepSeek V4 Pro scores higher for coding on the public lane, 51.4 to 36.7. GPT-5 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like. BenchLM does not name a winner for coding from a directional reading; read the shared benchmark rows directly and test the models on the same task set.

Which is better for agentic tasks, DeepSeek V4 Pro or GPT-5?

GPT-5 is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.

Which costs less, DeepSeek V4 Pro or GPT-5?

For the stated presets, chat costs $0.0033 on DeepSeek V4 Pro and $0.00625 on GPT-5; repository review costs $0.07788 and $0.0925; the cache-heavy agent loop costs $0.0748 and $0.375. GPT-5 has no published cached-input rate, so cached tokens use its listed input rate.

Which has the larger context window, DeepSeek V4 Pro or GPT-5?

DeepSeek V4 Pro has the larger documented context window: 1M, compared with 400K.

Self-host vs API cost

Estimates at 50,000 req/day · 1000 tokens/req average.

DeepSeek V4 Pro
API / mo$3,960
Self-host / mo$37,843
Break-evenNot reached at 70% load
GPT-5
API / mo$8,438
Self-host / moNot listed
Break-even—
Proprietary model — self-hosting not applicable.
Model the full break-even

Benchmark evidence

The full public result ledger is available for audit without forcing a wide desktop table onto a phone.

Browse raw public benchmark evidence46 rows

Agentic

  • Terminal-Bench 4.1

    DeepSeek V4 Pro6.70%
    Source
    GPT-5—

    Not directly comparable

  • Finch

    DeepSeek V4 Pro67.4%
    Source
    GPT-5—

    Not directly comparable

  • Terminal-Bench 2.0

    DeepSeek V4 Pro67.9%
    Source
    GPT-5—

    Not directly comparable

  • Terminal-Bench 2.1

    DeepSeek V4 Pro87.9%
    Source
    GPT-5—

    Not directly comparable

  • BrowseComp

    DeepSeek V4 Pro83.4%
    Source
    GPT-5—

    Not directly comparable

  • HLE w/ tools

    DeepSeek V4 Pro60.0%
    Source
    GPT-5—

    Not directly comparable

  • MCP Atlas

    DeepSeek V4 Pro73.6%
    Source
    GPT-5—

    Not directly comparable

  • Toolathlon

    DeepSeek V4 Pro51.8%
    Source
    GPT-5—

    Not directly comparable

  • CyberGym

    DeepSeek V4 Pro83.3%
    Source
    GPT-5—

    Not directly comparable

  • Toolathlon-Verified

    DeepSeek V4 Pro74.1%
    Source
    GPT-5—

    Not directly comparable

  • Agents' Last Exam

    DeepSeek V4 Pro25.7%
    Source
    GPT-5—

    Not directly comparable

  • AutomationBench

    DeepSeek V4 Pro31.8%
    Source
    GPT-5—

    Not directly comparable

  • Terminal-Bench 2.1 (Vals)

    DeepSeek V4 Pro54.7%
    Source
    GPT-5—

    Not directly comparable

  • JobBench

    DeepSeek V4 Pro—
    GPT-58.5%
    Source

    Not directly comparable

Coding

  • LiveCodeBench Pass@1-COT

    DeepSeek V4 Pro93.5%
    Source
    GPT-5—

    Not directly comparable

  • Codeforces

    DeepSeek V4 Pro3206.0
    Source
    GPT-5—

    Not directly comparable

  • SWE-bench Verified

    DeepSeek V4 Pro80.6%
    Source
    GPT-5—

    Not directly comparable

  • SWE-bench Pro

    DeepSeek V4 Pro55.4%
    Source
    GPT-5—

    Not directly comparable

  • SWE Multilingual

    DeepSeek V4 Pro76.2%
    Source
    GPT-5—

    Not directly comparable

  • Terminal-Bench 2.0

    DeepSeek V4 Pro67.9%
    Source
    GPT-5—

    Not directly comparable

  • Vibe Code Bench

    Shared source
    DeepSeek V4 Pro49.93%
    GPT-520.09%

    DeepSeek V4 Pro leads this result

  • Terminal-Bench 2.1

    DeepSeek V4 Pro87.9%
    Source
    GPT-5—

    Not directly comparable

  • NL2Repo

    DeepSeek V4 Pro61.5%
    Source
    GPT-5—

    Not directly comparable

  • DeepSWE

    DeepSeek V4 Pro62.7%
    Source
    GPT-5—

    Not directly comparable

  • DSBench-FullStack

    DeepSeek V4 Pro71.1%
    Source
    GPT-5—

    Not directly comparable

  • DSBench-Hard

    DeepSeek V4 Pro67.2%
    Source
    GPT-5—

    Not directly comparable

  • OpenHarmony Bench v1.0

    DeepSeek V4 Pro59.0%
    Source
    GPT-5—

    Not directly comparable

  • LiveCodeBench (Vals)

    DeepSeek V4 Pro87.5%
    Source
    GPT-585.9%
    Source

    DeepSeek V4 Pro leads this result

  • SWE-bench (Vals)

    DeepSeek V4 Pro96.4%
    Source
    GPT-569.0%
    Source

    DeepSeek V4 Pro leads this result

Reasoning

  • MRCR 1M

    DeepSeek V4 Pro83.5%
    Source
    GPT-5—

    Not directly comparable

  • CorpusQA 1M

    DeepSeek V4 Pro62.0%
    Source
    GPT-5—

    Not directly comparable

  • ARC-AGI-1

    DeepSeek V4 Pro90.00%
    Source
    GPT-5—

    Not directly comparable

  • ARC-AGI-2

    DeepSeek V4 Pro61.3%
    Source
    GPT-5—

    Not directly comparable

Knowledge

  • MMLU-Pro

    DeepSeek V4 Pro87.5%
    Source
    GPT-5—

    Not directly comparable

  • Chinese-SimpleQA

    DeepSeek V4 Pro84.4%
    Source
    GPT-5—

    Not directly comparable

  • GPQA

    DeepSeek V4 Pro90.1%
    Source
    GPT-5—

    Not directly comparable

  • GPQA-D

    DeepSeek V4 Pro90.1%
    Source
    GPT-5—

    Not directly comparable

  • HLE

    DeepSeek V4 Pro42.7%
    Source
    GPT-5—

    Not directly comparable

  • GPQA Diamond (Vals)

    DeepSeek V4 Pro92.4%
    Source
    GPT-585.6%
    Source

    DeepSeek V4 Pro leads this result

  • MMLU-Pro (Vals)

    DeepSeek V4 Pro87.0%
    Source
    GPT-586.5%
    Source

    DeepSeek V4 Pro leads this result

Math

  • HMMT Feb 2026

    DeepSeek V4 Pro95.2%
    Source
    GPT-5—

    Not directly comparable

  • IMOAnswerBench

    DeepSeek V4 Pro89.8%
    Source
    GPT-5—

    Not directly comparable

  • Apex

    DeepSeek V4 Pro38.3%
    Source
    GPT-5—

    Not directly comparable

  • Apex Shortlist

    DeepSeek V4 Pro90.2%
    Source
    GPT-5—

    Not directly comparable

  • FrontierMath v2 (Tiers 1-3)

    DeepSeek V4 Pro64.561%
    Source
    GPT-5—

    Not directly comparable

  • FrontierMath v2 (Tier 4)

    DeepSeek V4 Pro26.829%
    Source
    GPT-5—

    Not directly comparable

46 public results · 5 shared

Watch DeepSeek V4 Pro vs GPT-5

One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.

Read a sample issue

Join 7,000+ subscribers.

Last updated November 18, 2026