LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

GPT-4 vs Hy3 preview

Updated October 21, 2026. 28 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace GPT-4 Replace Hy3 preview
GPT-4OpenAI
Coding
22.4
Overall
6.7
Output $/M
$60.00
0
Hy3 previewTencent
Coding
22.4
Overall
25.3
Output $/M
$0.60
28
vs boards won of 28 shared

Decision reading

GPT-4 and Hy3 preview are level for coding, 22.4 against 22.4 on the Vibe Coding Index, so price and context decide.

Pick Hy3 preview when output tokens cost 100.0x less, or when the context window is larger, 262K against 8K tokens.

Model AGPT-4OpenAI22.4/1001 independent benchmarkNo cost per solved taskModel BHy3 previewTencent22.4/1001 independent benchmarkNo cost per solved task
Shared results
28
GPT-4 only
17
Hy3 preview only
8
Like-for-like categories
6 of 9

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

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
GPT-4 22.4 from 1 benchmark, Hy3 preview 22.4 from 1 benchmark.
Widest category
Reasoning, 41.8 points apart on average over 4 shared benchmarks.
Widest single gap
GPQA Diamond (Artificial Analysis run): GPT-4 34.9% against Hy3 preview 86.7%, 51.8 points apart on a 0 to 100 scale.
Output price
GPT-4 $60.00 against Hy3 preview $0.60 per million output tokens, 100.0x apart.
Context window
GPT-4 takes 8K tokens, Hy3 preview 262K.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryGPT-4BarsHy3 previewTies
Coding03
Intelligence01
Reasoning04
Math02
Knowledge04
General014
All shared boards028
CodingLike-for-likeFour of these boards feed the Vibe Coding Index03
Coding benchmarks for GPT-4 and Hy3 preview
BenchmarkGPT-4Hy3 previewLeader
SWE-bench Verified22.4%Not measuredOnly GPT-4 measured
SWE-bench Lite39.7%Not measuredOnly GPT-4 measured
SWE-bench (full test set)12.5%Not measuredOnly GPT-4 measured
LMArena WebDevNot measured1357Only Hy3 preview measured
AA Coding Index13.158.8Hy3 preview
Terminal-Bench Hard (Artificial Analysis run)Not measured34.1%Only Hy3 preview measured
LMArena Text: coding12091421Hy3 preview
LMArena Text: software and IT11951423Hy3 preview
LMArena WebDev: ReactNot measured1349Only Hy3 preview measured
LMArena WebDev: HTMLNot measured1356Only Hy3 preview measured
Neither model has a result on the other 28 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for GPT-4 and Hy3 preview
BenchmarkGPT-4Hy3 previewLeader
AA Intelligence Index6.725.3Hy3 preview
Epoch Capabilities Index125.7Not measuredOnly GPT-4 measured
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
Agentic work benchmarks for GPT-4 and Hy3 preview
BenchmarkGPT-4Hy3 previewLeader
AA Agentic IndexNot measured24.1Only Hy3 preview measured
τ²-Bench TelecomNot measured92.7%Only Hy3 preview measured
Neither model has a result on the other 37 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index04
Reasoning benchmarks for GPT-4 and Hy3 preview
BenchmarkGPT-4Hy3 previewLeader
GPQA Diamond (Artificial Analysis run)34.9%86.7%Hy3 preview
Humanity's Last Exam (Artificial Analysis run)Not measured27.8%Only Hy3 preview measured
BIG-Bench Hard75.1%Not measuredOnly GPT-4 measured
LMArena Text: hard prompts12001413Hy3 preview
LMArena Text: expert11491422Hy3 preview
LMArena Text: hard prompts in English12091416Hy3 preview
Neither model has a result on the other 15 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index02
Math benchmarks for GPT-4 and Hy3 preview
BenchmarkGPT-4Hy3 previewLeader
GSM8K92.0%Not measuredOnly GPT-4 measured
LMArena Text: math12301424Hy3 preview
LMArena Text: mathematical work12281442Hy3 preview
Neither model has a result on the other 10 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index04
Knowledge benchmarks for GPT-4 and Hy3 preview
BenchmarkGPT-4Hy3 previewLeader
MMLU Pro (Artificial Analysis run)56.2%Not measuredOnly GPT-4 measured
MMLU86.4%Not measuredOnly GPT-4 measured
AI2 Reasoning Challenge96.3%Not measuredOnly GPT-4 measured
LMArena Text: science11771416Hy3 preview
LMArena Text: medicine and healthcare11451405Hy3 preview
LMArena Text: legal and government11911430Hy3 preview
LMArena Text: business and finance11451397Hy3 preview
Neither model has a result on the other 10 boards here
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Web design benchmarks for GPT-4 and Hy3 preview
BenchmarkGPT-4Hy3 previewLeader
Neither model has a result on 9 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index014
General benchmarks for GPT-4 and Hy3 preview
BenchmarkGPT-4Hy3 previewLeader
IFBench33.2%63.1%Hy3 preview
Long context reasoning (AA-LCR)Not measured64.7%Only Hy3 preview measured
DTBench62.7%Not measuredOnly GPT-4 measured
LMCA17.1%Not measuredOnly GPT-4 measured
HellaSwag95.3%Not measuredOnly GPT-4 measured
WinoGrande87.5%Not measuredOnly GPT-4 measured
LMArena Text12061401Hy3 preview
LMArena Text: instruction following12011383Hy3 preview
LMArena Text: creative writing11911339Hy3 preview
LMArena Text: English12201409Hy3 preview
LMArena Text: non-English11941390Hy3 preview
LMArena Text: Chinese11841435Hy3 preview
LMArena Text: German1198Not measuredOnly GPT-4 measured
LMArena Text: Japanese1137Not measuredOnly GPT-4 measured
LMArena Text: Korean1087Not measuredOnly GPT-4 measured
LMArena Text: French12201425Hy3 preview
LMArena Text: Spanish11971399Hy3 preview
LMArena Text: Russian11951399Hy3 preview
LMArena Text: Polish1265Not measuredOnly GPT-4 measured
LMArena Text: multi-turn12061411Hy3 preview
LMArena Text: longer queries11901402Hy3 preview
LMArena Text: writing and language12191370Hy3 preview
LMArena Text: entertainment and media11841352Hy3 preview
Neither model has a result on the other 23 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-4 and Hy3 preview
BenchmarkGPT-4Hy3 previewLeader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

GPT-4$0.060Fits in one request
Hy3 preview$0.0005Fits in one request

Hy3 preview costs less

Repository review

50K input + 3K output tokens

GPT-4$1.68Exceeds the context window
Hy3 preview$0.011Fits in one request

Hy3 preview costs less

Agent loop

200K input + 10K output tokens

GPT-4$6.60Exceeds the context window
Hy3 preview$0.042Fits in one request

Hy3 preview costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of GPT-4 and Hy3 preview
SpecificationOpenAI GPT-4Tencent Hy3 preview
MakerOpenAITencent
WeightsNo public weights listedOpen weights
ReleasedMay 28, 2023April 22, 2026
Context window8K262K
Input $/Mtok$30.00$0.18
Output $/Mtok$60.00$0.60
Coding index22.422.4
Independent measurements behind it11
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-4 better than Hy3 preview for coding?

GPT-4 and Hy3 preview are level for coding, 22.4 against 22.4 on the Vibe Coding Index, so price and context decide. Pick Hy3 preview when output tokens cost 100.0x less, or when the context window is larger, 262K against 8K tokens.

Which is cheaper, GPT-4 or Hy3 preview?

GPT-4 costs $30.00 per million input tokens and $60.00 per million output tokens; Hy3 preview costs $0.18 and $0.60. Cost per solved task is known for neither of them.

Which has the bigger context window, GPT-4 or Hy3 preview?

Hy3 preview, with 262K tokens against 8K.

How many benchmarks is the GPT-4 vs Hy3 preview comparison based on?

28 benchmarks have a result for both models as of October 21, 2026. GPT-4's coding index rests on 1 independent benchmark and Hy3 preview's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next