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 the free Radar Brief
BenchLM
Model A
dots3-note Preview

Dots Studio

68.66/100

Estimated · Public rank #22

90% interval 58.8–78.5

dots3-note Preview vs Hy4 preview

Updated August 28, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

Tencent logo
Model B
Hy4 preview

Tencent

79.16/100

Estimated · Public rank #7

90% interval 69.3–89.0

Decision reading

Hy4 preview has the higher public score estimate, 79.16 versus 68.66, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.

11 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

    Hy4 preview

    Hy4 preview 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

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

    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: listed-rates

  • 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
11
dots3-note Preview only
20
Hy4 preview only
17
Like-for-like categories
0 / 8

2 categories use 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.

Coding

Directional only
dots3-note Preview
71.7
Hy4 preview
65.7
Weighted basis
2 vs 1 rows
Reading
Directional only

Knowledge

Directional only
dots3-note Preview
52.6
Hy4 preview
60.4
Weighted basis
1 vs 2 rows
Reading
Directional only

Agentic

Not comparable
dots3-note Preview
83.3
Hy4 preview
Not measured
Weighted basis
1 vs 0 rows
Reading
Not comparable

Reasoning

Not comparable
dots3-note Preview
81.4
Hy4 preview
Not measured
Weighted basis
1 vs 0 rows
Reading
Not comparable

Math

Not comparable
dots3-note Preview
Not measured
Hy4 preview
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multilingual

Not comparable
dots3-note Preview
Not measured
Hy4 preview
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multimodal

Not comparable
dots3-note Preview
79.1
Hy4 preview
66.2
Weighted basis
1 vs 1 rows
Reading
Not comparable

Instruction following

Not comparable
dots3-note Preview
85.1
Hy4 preview
Not measured
Weighted basis
2 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

dots3-note Preview
Self-hosted; infrastructure cost varies
Fits in one request
Hy4 preview
Self-hosted; infrastructure cost varies
Fits in one request

dots3-note Preview has no comparable published API token rate. Hy4 preview has no comparable published API token rate.

Repository review

50K fresh input + 3K output tokens

dots3-note Preview
Self-hosted; infrastructure cost varies
Fits in one request
Hy4 preview
Self-hosted; infrastructure cost varies
Fits in one request

dots3-note Preview has no comparable published API token rate. Hy4 preview has no comparable published API token rate.

Cache-heavy agent loop

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

dots3-note Preview
Self-hosted; infrastructure cost varies
Fits in one request
Cached-input rate unavailable
Hy4 preview
Self-hosted; infrastructure cost varies
Fits in one request
Cached-input rate unavailable

dots3-note Preview has no comparable published API token rate. Hy4 preview 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.

Documented inputs

dots3-note Preview

Not sourced

Hy4 preview

Not sourced

Documented outputs

dots3-note Preview

Not sourced

Hy4 preview

Not sourced

Provider availability

dots3-note Preview

Not sourced

Hy4 preview

Not sourced

Reasoning profile

dots3-note Preview

Reasoning

Hy4 preview

Reasoning

Weight access

dots3-note Preview

Open Weight

Hy4 preview

Open Weight

License

dots3-note Preview

Open Weight

Hy4 preview

Open Weight

Release date

dots3-note Preview

2026-08-14

Hy4 preview

2026-08-28

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
Hy4 preview has the higher public score estimate, 79.16 versus 68.66, but the 90% score intervals overlap.
Workload cost
A complete comparable API-rate estimate is not available for both models.
Context tradeoff
Hy4 preview 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 evidence48 rows

Agentic

  • Claw-Eval

    dots3-note Preview73.4%
    Source
    Hy4 preview—

    Not directly comparable

  • Terminal-Bench 2.1

    dots3-note Preview75.1%
    Source
    Hy4 preview85.4%
    Source

    Hy4 preview leads this result

  • Toolathlon-Verified

    dots3-note Preview55.6%
    Source
    Hy4 preview74.1%
    Source

    Hy4 preview leads this result

  • skillsBench

    dots3-note Preview52.8%
    Source
    Hy4 preview62.9%
    Source

    Hy4 preview leads this result

  • APEX-Agents

    dots3-note Preview30.8%
    Source
    Hy4 preview37.1%
    Source

    Hy4 preview leads this result

  • BrowseComp

    dots3-note Preview83.3%
    Source
    Hy4 preview—

    Not directly comparable

  • HLE w/ tools

    dots3-note Preview52.6%
    Source
    Hy4 preview55.4%
    Source

    Hy4 preview leads this result

  • DeepSearchQA

    dots3-note Preview92.1%
    Source
    Hy4 preview—

    Not directly comparable

  • WideResearch

    dots3-note Preview78.9%
    Source
    Hy4 preview83.9%
    Source

    Hy4 preview leads this result

  • CyberGym

    dots3-note Preview—
    Hy4 preview78.4%
    Source

    Not directly comparable

  • DRACO

    dots3-note Preview—
    Hy4 preview77.2%
    Source

    Not directly comparable

  • MCP Atlas

    dots3-note Preview—
    Hy4 preview83.7%
    Source

    Not directly comparable

  • JobBench

    dots3-note Preview—
    Hy4 preview61.7%
    Source

    Not directly comparable

  • Agents' Last Exam

    dots3-note Preview—
    Hy4 preview22.8%
    Source

    Not directly comparable

  • AutomationBench

    dots3-note Preview—
    Hy4 preview32.1%
    Source

    Not directly comparable

  • BankerToolBench

    dots3-note Preview—
    Hy4 preview78.6%
    Source

    Not directly comparable

Coding

  • Codeforces

    dots3-note Preview3056.0
    Source
    Hy4 preview—

    Not directly comparable

  • LiveCodeBench v6

    dots3-note Preview91.5%
    Source
    Hy4 preview—

    Not directly comparable

  • Terminal-Bench 2.1

    dots3-note Preview75.1%
    Source
    Hy4 preview85.4%
    Source

    Hy4 preview leads this result

  • SWE-bench Verified

    dots3-note Preview78.4%
    Source
    Hy4 preview—

    Not directly comparable

  • SWE Multilingual

    dots3-note Preview75.7%
    Source
    Hy4 preview82.9%
    Source

    Hy4 preview leads this result

  • SWE-bench Pro

    dots3-note Preview61%
    Source
    Hy4 preview65.7%
    Source

    Hy4 preview leads this result

  • NL2Repo

    dots3-note Preview49.8%
    Source
    Hy4 preview58.9%
    Source

    Hy4 preview leads this result

  • deepSwe

    dots3-note Preview—
    Hy4 preview64.3%
    Source

    Not directly comparable

  • ProgramBench

    dots3-note Preview—
    Hy4 preview17.5%
    Source

    Not directly comparable

  • PostTrain Bench

    dots3-note Preview—
    Hy4 preview35.6%
    Source

    Not directly comparable

  • sweMarathon

    dots3-note Preview—
    Hy4 preview31.9%
    Source

    Not directly comparable

Reasoning

  • ARC-AGI-2

    dots3-note Preview81.4%
    Source
    Hy4 preview—

    Not directly comparable

  • CritPt

    dots3-note Preview—
    Hy4 preview16.9%
    Source

    Not directly comparable

Knowledge

  • HLE

    dots3-note Preview52.6%
    Source
    Hy4 preview55.4%
    Source

    Hy4 preview leads this result

  • GPQA

    dots3-note Preview—
    Hy4 preview92.3%
    Source

    Not directly comparable

  • GPQA-D

    dots3-note Preview—
    Hy4 preview92.3%
    Source

    Not directly comparable

  • HLE w/o tools

    dots3-note Preview—
    Hy4 preview43.4%
    Source

    Not directly comparable

Math

  • IMOAnswerBench

    dots3-note Preview90.9%
    Source
    Hy4 preview—

    Not directly comparable

  • Apex

    dots3-note Preview—
    Hy4 preview74.2%
    Source

    Not directly comparable

Multimodal

  • SimpleVQA

    dots3-note Preview72.5%
    Source
    Hy4 preview—

    Not directly comparable

  • MMMU-Pro

    dots3-note Preview79.1%
    Source
    Hy4 preview—

    Not directly comparable

  • MathVision

    dots3-note Preview87.7%
    Source
    Hy4 preview—

    Not directly comparable

  • ZeroBench

    dots3-note Preview19.0%
    Source
    Hy4 preview—

    Not directly comparable

  • CharXiv w/o tools

    dots3-note Preview83.1%
    Source
    Hy4 preview—

    Not directly comparable

  • GDP.pdf (no tools)

    dots3-note Preview60.7%
    Source
    Hy4 preview—

    Not directly comparable

  • PerceptionBench

    dots3-note Preview53.4%
    Source
    Hy4 preview—

    Not directly comparable

  • BabyVision

    dots3-note Preview50.0%
    Source
    Hy4 preview—

    Not directly comparable

  • MMVU

    dots3-note Preview79.9%
    Source
    Hy4 preview—

    Not directly comparable

  • VideoMMMU

    dots3-note Preview86.8%
    Source
    Hy4 preview—

    Not directly comparable

  • OfficeQA Pro

    dots3-note Preview—
    Hy4 preview66.2%
    Source

    Not directly comparable

Instruction following

  • IFBench

    dots3-note Preview80.4%
    Source
    Hy4 preview—

    Not directly comparable

  • IFEval

    dots3-note Preview93.9%
    Source
    Hy4 preview—

    Not directly comparable

Frequently asked questions

Which is better, dots3-note Preview or Hy4 preview?

Hy4 preview has the higher public score estimate, 79.16 versus 68.66, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.

Which is better for coding, dots3-note Preview or Hy4 preview?

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, dots3-note Preview or Hy4 preview?

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, dots3-note Preview or Hy4 preview?

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, dots3-note Preview or Hy4 preview?

Hy4 preview has the larger documented context window: 1M, compared with 512K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated August 28, 2026

Watch dots3-note Preview vs Hy4 preview

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

Read a sample issue

Join 2,000+ readers.