Skip to main content
BenchLM
Advertise

Model comparison

DeepSeek V4 Pro Base vs Pokee-Isaac 28B

Updated August 4, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

DeepSeek V4 Pro Base

DeepSeek

—

Evidence status unavailable

90% interval unavailable

Pokee-Isaac 28B

Pokee AI

—

Evidence status unavailable

90% interval unavailable

The public evidence has no benchmark result shared by both models, so it does not support a quality verdict. Use the documented cost, context, and runtime rows instead.

0 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.

  • Long documents

    Prompts that approach the documented context limit

    Pokee-Isaac 28B

    Pokee-Isaac 28B has the larger documented context window.

    Confidence: documented

Show secondary and unsupported calls

  • Coding work

    Code generation, repair, and software-engineering tasks

    Not enough matched evidence

    No shared weighted benchmark basis supports a winner.

    Confidence: limited

  • Agentic work

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

    Not enough matched evidence

    No shared weighted benchmark basis supports a winner.

    Confidence: limited

  • Chat turn cost

    1K fresh input + 500 output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates

  • Cache-heavy agent loop cost

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

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: rate-fallback

  • Repository review cost

    50K fresh input + 3K output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates

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
0
DeepSeek V4 Pro Base only
24
Pokee-Isaac 28B only
7
Like-for-like categories
0 / 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

Not comparable
DeepSeek V4 Pro Base
Not measured
Pokee-Isaac 28B
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Coding

Not comparable
DeepSeek V4 Pro Base
Not measured
Pokee-Isaac 28B
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Reasoning

Not comparable
DeepSeek V4 Pro Base
51.5
Pokee-Isaac 28B
60.7
Weighted basis
1 vs 1 rows
Reading
Not comparable

Knowledge

Not comparable
DeepSeek V4 Pro Base
66.4
Pokee-Isaac 28B
Not measured
Weighted basis
3 vs 0 rows
Reading
Not comparable

Math

Not comparable
DeepSeek V4 Pro Base
Not measured
Pokee-Isaac 28B
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multilingual

Not comparable
DeepSeek V4 Pro Base
Not measured
Pokee-Isaac 28B
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multimodal

Not comparable
DeepSeek V4 Pro Base
Not measured
Pokee-Isaac 28B
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Instruction following

Not comparable
DeepSeek V4 Pro Base
Not measured
Pokee-Isaac 28B
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.

A shared-evidence shape is not available.

BenchLM does not draw a radar or infer missing axes when the matched evidence is too sparse.

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 Base
API rate not published
Fits in one request
Pokee-Isaac 28B
$0.00065
Fits in one request

DeepSeek V4 Pro Base has no comparable published API token rate.

Repository review

50K fresh input + 3K output tokens

DeepSeek V4 Pro Base
API rate not published
Fits in one request
Pokee-Isaac 28B
$0.0105
Fits in one request

DeepSeek V4 Pro Base has no comparable published API token rate.

Cache-heavy agent loop

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

DeepSeek V4 Pro Base
API rate not published
Fits in one request
Cached-input rate unavailable
Pokee-Isaac 28B
$0.043
Fits in one request
Cached input priced at the published list-input rate

Pokee-Isaac 28B has no published cached-input rate, so cached tokens use its listed input rate. DeepSeek V4 Pro Base has no comparable published API token rate.

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 Base

No comparable hosted API rate

Pokee-Isaac 28B

Not published

Pokee-Isaac 28B model page

Reasoning profile

DeepSeek V4 Pro Base

Non-Reasoning

Pokee-Isaac 28B

Reasoning

Weight access

DeepSeek V4 Pro Base

Open Weight

Pokee-Isaac 28B

Proprietary

License

DeepSeek V4 Pro Base

Open Weight

Pokee-Isaac 28B

Proprietary

Release date

DeepSeek V4 Pro Base

2026-04-24

Pokee-Isaac 28B

2026-08-03

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
The public evidence has no benchmark result shared by both models, so it does not support a quality verdict.
Workload cost
A complete comparable API-rate estimate is not available for both models.
Context tradeoff
Pokee-Isaac 28B has the larger documented window (10M).

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 evidence31 rows

Agentic

  • Terminal-Bench 2.1

    DeepSeek V4 Pro Base—
    Pokee-Isaac 28B65.1%
    Source

    Not directly comparable

  • BFCL v4

    DeepSeek V4 Pro Base—
    Pokee-Isaac 28B70.9%
    Source

    Not directly comparable

  • τ³-bench results

    DeepSeek V4 Pro Base—
    Pokee-Isaac 28B66.2%
    Source

    Not directly comparable

  • MCP-Atlas claim coverage

    DeepSeek V4 Pro Base—
    Pokee-Isaac 28B74.6%
    Source

    Not directly comparable

  • PinchBench

    DeepSeek V4 Pro Base—
    Pokee-Isaac 28B95.7%
    Source

    Not directly comparable

Coding

  • BigCodeBench

    DeepSeek V4 Pro Base59.2%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • HumanEval

    DeepSeek V4 Pro Base76.8%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • Terminal-Bench 2.1

    DeepSeek V4 Pro Base—
    Pokee-Isaac 28B65.1%
    Source

    Not directly comparable

Reasoning

  • BBH

    DeepSeek V4 Pro Base87.5%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • DROP

    DeepSeek V4 Pro Base88.7%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • HellaSwag

    DeepSeek V4 Pro Base88.0%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • WinoGrande

    DeepSeek V4 Pro Base81.5%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • CLUEWSC

    DeepSeek V4 Pro Base85.2%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • LongBench v2

    DeepSeek V4 Pro Base51.5%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • MRCRv2

    DeepSeek V4 Pro Base—
    Pokee-Isaac 28B60.7%
    Source

    Not directly comparable

Knowledge

  • AGIEval

    DeepSeek V4 Pro Base83.1%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • MMLU

    DeepSeek V4 Pro Base90.1%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • MMLU-Redux

    DeepSeek V4 Pro Base90.8%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • MMLU-Pro

    DeepSeek V4 Pro Base73.5%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • MMMLU

    DeepSeek V4 Pro Base90.3%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • C-Eval

    DeepSeek V4 Pro Base93.1%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • CMMLU

    DeepSeek V4 Pro Base90.8%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • MultiLoKo

    DeepSeek V4 Pro Base51.1%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • SimpleQA

    DeepSeek V4 Pro Base55.2%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • SuperGPQA

    DeepSeek V4 Pro Base53.9%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • FACTS Parametric

    DeepSeek V4 Pro Base62.6%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • TriviaQA

    DeepSeek V4 Pro Base85.6%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

Math

  • GSM8K

    DeepSeek V4 Pro Base92.6%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • MATH

    DeepSeek V4 Pro Base64.5%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

  • CMath

    DeepSeek V4 Pro Base90.9%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

Multilingual

  • MGSM

    DeepSeek V4 Pro Base84.4%
    Source
    Pokee-Isaac 28B—

    Not directly comparable

Frequently asked questions

Which is better, DeepSeek V4 Pro Base or Pokee-Isaac 28B?

The public evidence has no benchmark result shared by both models, so it does not support a quality verdict. The page therefore keeps the decision tied to the specific documented workload.

Which is better for coding, DeepSeek V4 Pro Base or Pokee-Isaac 28B?

The published evidence does not provide a shared weighted coding basis for both models, so BenchLM does not name a coding winner.

Which is better for agentic tasks, DeepSeek V4 Pro Base or Pokee-Isaac 28B?

The published evidence does not provide a shared weighted agentic tasks basis for both models, so BenchLM does not name a agentic tasks winner.

Which costs less, DeepSeek V4 Pro Base or Pokee-Isaac 28B?

Both models do not have comparable published API token rates, so this page does not name a universal price winner.

Which has the larger context window, DeepSeek V4 Pro Base or Pokee-Isaac 28B?

Pokee-Isaac 28B has the larger documented context window: 10M, compared with 1M.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated August 4, 2026

Watch DeepSeek V4 Pro Base vs Pokee-Isaac 28B

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

Read a sample issue

Join 2,000+ readers.