Skip to main content
Radar

Five or fewer confirmed AI changes, with original sources, on mornings when something changed.A free source-linked morning brief.

Start free brief
BenchLM
Model A
DeepSeek V4 Flash 0731

DeepSeek

—

Evidence status unavailable

90% interval unavailable

DeepSeek V4 Flash 0731 vs DeepSeek V4 Pro 0813

Updated August 13, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

Model B
DeepSeek V4 Pro 0813

DeepSeek

60.9/100

Estimated · Public rank #48

90% interval 51.1–70.8

Decision reading

At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.

33 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.

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.

  • Coding work

    Code generation, repair, and software-engineering tasks

    DeepSeek V4 Pro 0813

    DeepSeek V4 Pro 0813 leads on the same 2 weighted benchmark rows.

    Confidence: limited

  • Agentic work

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

    DeepSeek V4 Pro 0813

    DeepSeek V4 Pro 0813 leads on the same 2 weighted benchmark rows.

    Confidence: limited

  • Chat turn cost

    1K fresh input + 500 output tokens

    DeepSeek V4 Flash 0731

    DeepSeek V4 Flash 0731 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates

Show secondary and unsupported calls

  • Cache-heavy agent loop cost

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

    DeepSeek V4 Flash 0731

    DeepSeek V4 Flash 0731 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates

  • Repository review cost

    50K fresh input + 3K output tokens

    DeepSeek V4 Flash 0731

    DeepSeek V4 Flash 0731 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates

  • Long documents

    Prompts that approach the documented context limit

    No clear pick

    The documented context windows are equal.

    Confidence: documented

What is actually comparable

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

Shared results
33
DeepSeek V4 Flash 0731 only
0
DeepSeek V4 Pro 0813 only
1
Like-for-like categories
4 / 8

Category results, on a stated basis

Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.

Agentic

Like-for-like
DeepSeek V4 Flash 0731
63.8
DeepSeek V4 Pro 0813
74.5
Weighted basis
2 vs 2 rows
Reading
DeepSeek V4 Pro 0813 leads

Coding

Like-for-like
DeepSeek V4 Flash 0731
68.8
DeepSeek V4 Pro 0813
70.9
Weighted basis
2 vs 2 rows
Reading
DeepSeek V4 Pro 0813 leads

Knowledge

Like-for-like
DeepSeek V4 Flash 0731
55.3
DeepSeek V4 Pro 0813
62.5
Weighted basis
4 vs 4 rows
Reading
DeepSeek V4 Pro 0813 leads

Math

Like-for-like
DeepSeek V4 Flash 0731
94.8
DeepSeek V4 Pro 0813
95.2
Weighted basis
1 vs 1 rows
Reading
DeepSeek V4 Pro 0813 leads

Reasoning

Not comparable
DeepSeek V4 Flash 0731
Not measured
DeepSeek V4 Pro 0813
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multilingual

Not comparable
DeepSeek V4 Flash 0731
Not measured
DeepSeek V4 Pro 0813
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multimodal

Not comparable
DeepSeek V4 Flash 0731
Not measured
DeepSeek V4 Pro 0813
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Instruction following

Not comparable
DeepSeek V4 Flash 0731
Not measured
DeepSeek V4 Pro 0813
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

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.

  • SimpleQA

    Knowledge

    DeepSeek V4 Flash 0731: 34.1%DeepSeek V4 Pro 0813: 57.9%Normalized gap 23.8Shared source

  • Terminal-Bench 2.0

    Agentic

    DeepSeek V4 Flash 0731: 56.9%DeepSeek V4 Pro 0813: 67.9%Normalized gap 11.0Shared source

  • BrowseComp

    Agentic

    DeepSeek V4 Flash 0731: 73.2%DeepSeek V4 Pro 0813: 83.4%Normalized gap 10.2Shared source

  • HLE

    Knowledge

    DeepSeek V4 Flash 0731: 34.8%DeepSeek V4 Pro 0813: 42.7%Normalized gap 7.9DeepSeek V4 Flash 0731 source DeepSeek V4 Pro 0813 source

  • SWE-bench Pro

    Coding

    DeepSeek V4 Flash 0731: 52.6%DeepSeek V4 Pro 0813: 55.4%Normalized gap 2.8Shared source

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 Flash 0731
$0.00028
Fits in one request
DeepSeek V4 Pro 0813
$0.00087
Fits in one request

DeepSeek V4 Flash 0731 has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

DeepSeek V4 Flash 0731
$0.00784
Fits in one request
DeepSeek V4 Pro 0813
$0.02436
Fits in one request

DeepSeek V4 Flash 0731 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 Flash 0731
$0.00616
Fits in one request
DeepSeek V4 Pro 0813
$0.01812
Fits in one request

DeepSeek V4 Flash 0731 has the lower modeled cost

Costs use the listed standard API rates.

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 Flash 0731

$0.0028 per 1M cached input tokens

DeepSeek V4 Pro 0813

$0.003625 per 1M cached input tokens

Reasoning profile

DeepSeek V4 Flash 0731

Reasoning

DeepSeek V4 Pro 0813

Reasoning

Weight access

DeepSeek V4 Flash 0731

Proprietary

DeepSeek V4 Pro 0813

Proprietary

License

DeepSeek V4 Flash 0731

Proprietary

DeepSeek V4 Pro 0813

Proprietary

Release date

DeepSeek V4 Flash 0731

2026-07-31

DeepSeek V4 Pro 0813

2026-08-13

If you already use one of these models
Deployment change
Both entries list DeepSeek as the provider. Confirm endpoint, model ID, limits, and feature support before switching.
Quality signal
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
Workload cost
Repository review: $0.00784 vs $0.02436. Cache-heavy agent loop: $0.00616 vs $0.01812.
Context tradeoff
Both models list 1M.

Run the same representative tasks against both endpoints before changing production traffic.

Benchmark evidence

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

Browse raw public benchmark evidence34 rows

Agentic

  • Terminal-Bench 2.0

    Shared source
    DeepSeek V4 Flash 073156.9%
    DeepSeek V4 Pro 081367.9%

    DeepSeek V4 Pro 0813 leads this result

  • BrowseComp

    Shared source
    DeepSeek V4 Flash 073173.2%
    DeepSeek V4 Pro 081383.4%

    DeepSeek V4 Pro 0813 leads this result

  • HLE w/ tools

    DeepSeek V4 Flash 073145.1%
    Source
    DeepSeek V4 Pro 081360.0%
    Source

    DeepSeek V4 Pro 0813 leads this result

  • MCP Atlas

    Shared source
    DeepSeek V4 Flash 073169%
    DeepSeek V4 Pro 081373.6%

    DeepSeek V4 Pro 0813 leads this result

  • Toolathlon

    Shared source
    DeepSeek V4 Flash 073147.8%
    DeepSeek V4 Pro 081351.8%

    DeepSeek V4 Pro 0813 leads this result

  • Terminal-Bench 2.1

    DeepSeek V4 Flash 073182.7%
    Source
    DeepSeek V4 Pro 081387.9%
    Source

    DeepSeek V4 Pro 0813 leads this result

  • CyberGym

    DeepSeek V4 Flash 073176.7%
    Source
    DeepSeek V4 Pro 081383.3%
    Source

    DeepSeek V4 Pro 0813 leads this result

  • Toolathlon-Verified

    DeepSeek V4 Flash 073170.3%
    Source
    DeepSeek V4 Pro 081374.1%
    Source

    DeepSeek V4 Pro 0813 leads this result

  • Agents' Last Exam

    DeepSeek V4 Flash 073125.2%
    Source
    DeepSeek V4 Pro 081325.7%
    Source

    DeepSeek V4 Pro 0813 leads this result

  • AutomationBench

    DeepSeek V4 Flash 073125.1%
    Source
    DeepSeek V4 Pro 081331.8%
    Source

    DeepSeek V4 Pro 0813 leads this result

Coding

  • LiveCodeBench Pass@1-COT

    Shared source
    DeepSeek V4 Flash 073191.6%
    DeepSeek V4 Pro 081393.5%

    DeepSeek V4 Pro 0813 leads this result

  • Codeforces

    Shared source
    DeepSeek V4 Flash 07313052.0
    DeepSeek V4 Pro 08133206.0

    DeepSeek V4 Pro 0813 leads this result

  • SWE-bench Verified

    Shared source
    DeepSeek V4 Flash 073179%
    DeepSeek V4 Pro 081380.6%

    DeepSeek V4 Pro 0813 leads this result

  • SWE-bench Pro

    Shared source
    DeepSeek V4 Flash 073152.6%
    DeepSeek V4 Pro 081355.4%

    DeepSeek V4 Pro 0813 leads this result

  • SWE Multilingual

    Shared source
    DeepSeek V4 Flash 073173.3%
    DeepSeek V4 Pro 081376.2%

    DeepSeek V4 Pro 0813 leads this result

  • Terminal-Bench 2.0

    Shared source
    DeepSeek V4 Flash 073156.9%
    DeepSeek V4 Pro 081367.9%

    DeepSeek V4 Pro 0813 leads this result

  • Terminal-Bench 2.1

    DeepSeek V4 Flash 073182.7%
    Source
    DeepSeek V4 Pro 081387.9%
    Source

    DeepSeek V4 Pro 0813 leads this result

  • NL2Repo

    DeepSeek V4 Flash 073154.2%
    Source
    DeepSeek V4 Pro 081361.5%
    Source

    DeepSeek V4 Pro 0813 leads this result

  • deepSwe

    DeepSeek V4 Flash 073154.4%
    Source
    DeepSeek V4 Pro 081362.7%
    Source

    DeepSeek V4 Pro 0813 leads this result

  • DSBench-FullStack

    DeepSeek V4 Flash 073168.7%
    Source
    DeepSeek V4 Pro 081371.1%
    Source

    DeepSeek V4 Pro 0813 leads this result

  • DSBench-Hard

    DeepSeek V4 Flash 073159.6%
    Source
    DeepSeek V4 Pro 081367.2%
    Source

    DeepSeek V4 Pro 0813 leads this result

  • Vibe Code Bench

    DeepSeek V4 Flash 0731—
    DeepSeek V4 Pro 081349.93%
    Source

    Not directly comparable

Reasoning

  • MRCR 1M

    Shared source
    DeepSeek V4 Flash 073178.7%
    DeepSeek V4 Pro 081383.5%

    DeepSeek V4 Pro 0813 leads this result

  • CorpusQA 1M

    Shared source
    DeepSeek V4 Flash 073160.5%
    DeepSeek V4 Pro 081362.0%

    DeepSeek V4 Pro 0813 leads this result

Knowledge

  • MMLU-Pro

    Shared source
    DeepSeek V4 Flash 073186.2%
    DeepSeek V4 Pro 081387.5%

    DeepSeek V4 Pro 0813 leads this result

  • SimpleQA

    Shared source
    DeepSeek V4 Flash 073134.1%
    DeepSeek V4 Pro 081357.9%

    DeepSeek V4 Pro 0813 leads this result

  • Chinese-SimpleQA

    Shared source
    DeepSeek V4 Flash 073178.9%
    DeepSeek V4 Pro 081384.4%

    DeepSeek V4 Pro 0813 leads this result

  • GPQA

    Shared source
    DeepSeek V4 Flash 073188.1%
    DeepSeek V4 Pro 081390.1%

    DeepSeek V4 Pro 0813 leads this result

  • GPQA-D

    Shared source
    DeepSeek V4 Flash 073188.1%
    DeepSeek V4 Pro 081390.1%

    DeepSeek V4 Pro 0813 leads this result

  • HLE

    DeepSeek V4 Flash 073134.8%
    Source
    DeepSeek V4 Pro 081342.7%
    Source

    DeepSeek V4 Pro 0813 leads this result

Math

  • HMMT Feb 2026

    Shared source
    DeepSeek V4 Flash 073194.8%
    DeepSeek V4 Pro 081395.2%

    DeepSeek V4 Pro 0813 leads this result

  • IMOAnswerBench

    Shared source
    DeepSeek V4 Flash 073188.4%
    DeepSeek V4 Pro 081389.8%

    DeepSeek V4 Pro 0813 leads this result

  • Apex

    Shared source
    DeepSeek V4 Flash 073133.0%
    DeepSeek V4 Pro 081338.3%

    DeepSeek V4 Pro 0813 leads this result

  • Apex Shortlist

    Shared source
    DeepSeek V4 Flash 073185.7%
    DeepSeek V4 Pro 081390.2%

    DeepSeek V4 Pro 0813 leads this result

Frequently asked questions

Which is better, DeepSeek V4 Flash 0731 or DeepSeek V4 Pro 0813?

At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner. The page therefore keeps the decision tied to the specific documented workload.

Which is better for coding, DeepSeek V4 Flash 0731 or DeepSeek V4 Pro 0813?

DeepSeek V4 Pro 0813 leads the like-for-like coding comparison across 2 shared weighted benchmark rows.

Which is better for agentic tasks, DeepSeek V4 Flash 0731 or DeepSeek V4 Pro 0813?

DeepSeek V4 Pro 0813 leads the like-for-like agentic tasks comparison across 2 shared weighted benchmark rows.

Which costs less, DeepSeek V4 Flash 0731 or DeepSeek V4 Pro 0813?

For the stated presets, chat costs $0.00028 on DeepSeek V4 Flash 0731 and $0.00087 on DeepSeek V4 Pro 0813; repository review costs $0.00784 and $0.02436; the cache-heavy agent loop costs $0.00616 and $0.01812. Costs use the listed standard API rates.

Which has the larger context window, DeepSeek V4 Flash 0731 or DeepSeek V4 Pro 0813?

Both models list the same context window, 1M.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated August 13, 2026

Watch DeepSeek V4 Flash 0731 vs DeepSeek V4 Pro 0813

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

Read a sample issue

Join 2,000+ readers.