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 Pro 0813

DeepSeek

60.9/100

Estimated · Public rank #48

90% interval 51.1–70.8

DeepSeek V4 Pro 0813 vs Step 3.7 Flash

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

Model B
Step 3.7 Flash

StepFun

50.0/100

Estimated · Public rank #121

90% interval 38.5–61.5

Decision reading

DeepSeek V4 Pro 0813 has the higher public score estimate, 60.93 versus 49.96, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.

6 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.

  • 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

  • Long documents

    Prompts that approach the documented context limit

    DeepSeek V4 Pro 0813

    DeepSeek V4 Pro 0813 has the larger documented context window.

    Confidence: documented

  • Chat turn cost

    1K fresh input + 500 output tokens

    Step 3.7 Flash

    Step 3.7 Flash 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 Pro 0813

    DeepSeek V4 Pro 0813 has the lower estimated token cost for this stated workload. Step 3.7 Flash has no published cached-input rate, so cached tokens use its listed input rate.

    Confidence: rate-fallback

  • Repository review cost

    50K fresh input + 3K output tokens

    Step 3.7 Flash

    Step 3.7 Flash 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

    The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.

    Confidence: limited

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
6
DeepSeek V4 Pro 0813 only
28
Step 3.7 Flash only
5
Like-for-like categories
1 / 8

1 category uses different evidence sets. Those rows remain visible for coverage context but do not name a winner.

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 Pro 0813
74.5
Step 3.7 Flash
66.4
Weighted basis
2 vs 2 rows
Reading
DeepSeek V4 Pro 0813 leads

Coding

Directional only
DeepSeek V4 Pro 0813
70.9
Step 3.7 Flash
56.3
Weighted basis
2 vs 1 rows
Reading
Directional only

Reasoning

Not comparable
DeepSeek V4 Pro 0813
Not measured
Step 3.7 Flash
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Knowledge

Not comparable
DeepSeek V4 Pro 0813
62.5
Step 3.7 Flash
Not measured
Weighted basis
4 vs 0 rows
Reading
Not comparable

Math

Not comparable
DeepSeek V4 Pro 0813
95.2
Step 3.7 Flash
Not measured
Weighted basis
1 vs 0 rows
Reading
Not comparable

Multilingual

Not comparable
DeepSeek V4 Pro 0813
Not measured
Step 3.7 Flash
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multimodal

Not comparable
DeepSeek V4 Pro 0813
Not measured
Step 3.7 Flash
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Instruction following

Not comparable
DeepSeek V4 Pro 0813
Not measured
Step 3.7 Flash
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.

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 0813
$0.00087
Fits in one request
Step 3.7 Flash
$0.00077
Fits in one request

Step 3.7 Flash has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

DeepSeek V4 Pro 0813
$0.02436
Fits in one request
Step 3.7 Flash
$0.01345
Fits in one request

Step 3.7 Flash 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 0813
$0.01812
Fits in one request
Step 3.7 Flash
$0.0555
Fits in one request
Cached input priced at the published list-input rate

DeepSeek V4 Pro 0813 has the lower modeled cost

Step 3.7 Flash has no published cached-input rate, so cached tokens use its listed input rate.

Specification differences

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

Context window

Maximum documented context; output-token limits may be lower.

DeepSeek V4 Pro 0813

1M

DeepSeek models and pricing

Step 3.7 Flash

256K

Cached-input rate

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

DeepSeek V4 Pro 0813

$0.003625 per 1M cached input tokens

Step 3.7 Flash

Not published

Reasoning profile

DeepSeek V4 Pro 0813

Reasoning

Step 3.7 Flash

Reasoning

Weight access

DeepSeek V4 Pro 0813

Proprietary

Step 3.7 Flash

Open Weight

License

DeepSeek V4 Pro 0813

Proprietary

Step 3.7 Flash

Open Weight

Release date

DeepSeek V4 Pro 0813

2026-08-13

Step 3.7 Flash

2026-05-29

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 0813 has the higher public score estimate, 60.93 versus 49.96, but the 90% score intervals overlap.
Workload cost
Repository review: $0.02436 vs $0.01345. Cache-heavy agent loop: $0.01812 vs $0.0555.
Context tradeoff
DeepSeek V4 Pro 0813 has the larger documented window (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 evidence39 rows

Agentic

  • Terminal-Bench 2.0

    DeepSeek V4 Pro 081367.9%
    Source
    Step 3.7 Flash59.5%
    Source

    DeepSeek V4 Pro 0813 leads this result

  • Terminal-Bench 2.1

    DeepSeek V4 Pro 081387.9%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • BrowseComp

    DeepSeek V4 Pro 081383.4%
    Source
    Step 3.7 Flash75.8%
    Source

    DeepSeek V4 Pro 0813 leads this result

  • HLE w/ tools

    DeepSeek V4 Pro 081360.0%
    Source
    Step 3.7 Flash47.2%
    Source

    DeepSeek V4 Pro 0813 leads this result

  • MCP Atlas

    DeepSeek V4 Pro 081373.6%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • Toolathlon

    DeepSeek V4 Pro 081351.8%
    Source
    Step 3.7 Flash49.5%
    Source

    DeepSeek V4 Pro 0813 leads this result

  • CyberGym

    DeepSeek V4 Pro 081383.3%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • Toolathlon-Verified

    DeepSeek V4 Pro 081374.1%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • Agents' Last Exam

    DeepSeek V4 Pro 081325.7%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • AutomationBench

    DeepSeek V4 Pro 081331.8%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • DeepSearchQA

    DeepSeek V4 Pro 0813—
    Step 3.7 Flash92.8%
    Source

    Not directly comparable

  • Claw-Eval

    DeepSeek V4 Pro 0813—
    Step 3.7 Flash67.1%
    Source

    Not directly comparable

  • Gert Labs

    DeepSeek V4 Pro 0813—
    Step 3.7 Flash51.57%
    Source

    Not directly comparable

Coding

  • LiveCodeBench Pass@1-COT

    DeepSeek V4 Pro 081393.5%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • Codeforces

    DeepSeek V4 Pro 08133206.0
    Source
    Step 3.7 Flash—

    Not directly comparable

  • SWE-bench Verified

    DeepSeek V4 Pro 081380.6%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • SWE-bench Pro

    DeepSeek V4 Pro 081355.4%
    Source
    Step 3.7 Flash56.3%
    Source

    Step 3.7 Flash leads this result

  • SWE Multilingual

    DeepSeek V4 Pro 081376.2%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • Terminal-Bench 2.0

    DeepSeek V4 Pro 081367.9%
    Source
    Step 3.7 Flash59.5%
    Source

    DeepSeek V4 Pro 0813 leads this result

  • Vibe Code Bench

    DeepSeek V4 Pro 081349.93%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • Terminal-Bench 2.1

    DeepSeek V4 Pro 081387.9%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • NL2Repo

    DeepSeek V4 Pro 081361.5%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • deepSwe

    DeepSeek V4 Pro 081362.7%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • DSBench-FullStack

    DeepSeek V4 Pro 081371.1%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • DSBench-Hard

    DeepSeek V4 Pro 081367.2%
    Source
    Step 3.7 Flash—

    Not directly comparable

Reasoning

  • MRCR 1M

    DeepSeek V4 Pro 081383.5%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • CorpusQA 1M

    DeepSeek V4 Pro 081362.0%
    Source
    Step 3.7 Flash—

    Not directly comparable

Knowledge

  • MMLU-Pro

    DeepSeek V4 Pro 081387.5%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • SimpleQA

    DeepSeek V4 Pro 081357.9%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • Chinese-SimpleQA

    DeepSeek V4 Pro 081384.4%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • GPQA

    DeepSeek V4 Pro 081390.1%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • GPQA-D

    DeepSeek V4 Pro 081390.1%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • HLE

    DeepSeek V4 Pro 081342.7%
    Source
    Step 3.7 Flash—

    Not directly comparable

Math

  • HMMT Feb 2026

    DeepSeek V4 Pro 081395.2%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • IMOAnswerBench

    DeepSeek V4 Pro 081389.8%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • Apex

    DeepSeek V4 Pro 081338.3%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • Apex Shortlist

    DeepSeek V4 Pro 081390.2%
    Source
    Step 3.7 Flash—

    Not directly comparable

Multimodal

  • SimpleVQA

    DeepSeek V4 Pro 0813—
    Step 3.7 Flash79.2%
    Source

    Not directly comparable

  • V*

    DeepSeek V4 Pro 0813—
    Step 3.7 Flash95.3%
    Source

    Not directly comparable

Frequently asked questions

Which is better, DeepSeek V4 Pro 0813 or Step 3.7 Flash?

DeepSeek V4 Pro 0813 has the higher public score estimate, 60.93 versus 49.96, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.

Which is better for coding, DeepSeek V4 Pro 0813 or Step 3.7 Flash?

The current coding averages use different weighted benchmark sets, so BenchLM does not name a winner from them. Read the shared benchmark rows directly and test the models on the same task set.

Which is better for agentic tasks, DeepSeek V4 Pro 0813 or Step 3.7 Flash?

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

Which costs less, DeepSeek V4 Pro 0813 or Step 3.7 Flash?

For the stated presets, chat costs $0.00087 on DeepSeek V4 Pro 0813 and $0.00077 on Step 3.7 Flash; repository review costs $0.02436 and $0.01345; the cache-heavy agent loop costs $0.01812 and $0.0555. Step 3.7 Flash has no published cached-input rate, so cached tokens use its listed input rate.

Which has the larger context window, DeepSeek V4 Pro 0813 or Step 3.7 Flash?

DeepSeek V4 Pro 0813 has the larger documented context window: 1M, compared with 256K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated August 13, 2026

Watch DeepSeek V4 Pro 0813 vs Step 3.7 Flash

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

Read a sample issue

Join 2,000+ readers.