Enter opens the highlighted result. Esc closes.

DeepSeek V4 Pro 0813 vs Hy4 flash

DeepSeek V4 Pro 0813 leads Hy4 flash on coding, 56.1 to 44.5 on the Vibe Coding Index, and Hy4 flash costs 28.2x less per output token ($0.1405 vs $3.96 per million). Updated November 22, 2026; 7 benchmarks have a result for both.

DeepSeek V4 Pro 0813DeepSeek
Coding
56.1
Overall
61.8
Output $/M
$3.96
6
Hy4 flashTencent
Coding
44.5
Overall
No data
Output $/M
$0.1405
1
vs boards won of 7 shared

DeepSeek V4 Pro 0813 or Hy4 flash: which to pick

Pick DeepSeek V4 Pro 0813 for coding: it scores 56.1 on the Vibe Coding Index against 44.5 for Hy4 flash.

Pick Hy4 flash for output tokens 28.2x cheaper ($0.1405 vs $3.96 per million) and work measured by Terminal-Bench 4 (Vals AI), where it leads.

Shared results
7
DeepSeek V4 Pro 0813 only
78
Hy4 flash only
5
Like-for-like categories
3 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

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
DeepSeek V4 Pro 0813 56.1 from 5 benchmarks, Hy4 flash 44.5 from 2 benchmarks.
Widest category
Coding, 3.8 points apart on average over 3 shared benchmarks.
Widest single gap
Vibe Code Bench (Vals AI): DeepSeek V4 Pro 0813 82.3% against Hy4 flash 74.8%, 7.5 points apart on a 0 to 100 scale.
Output price
DeepSeek V4 Pro 0813 $3.96 against Hy4 flash $0.1405 per million output tokens, 28.2x apart.
Context window
Both take 1.05M tokens.

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
CategoryDeepSeek V4 Pro 0813BarsHy4 flashTies
Coding21
Agentic work30
General10
All shared boards61
CodingLike-for-likeSix of these boards feed the Vibe Coding Index21
Coding benchmarks for DeepSeek V4 Pro 0813 and Hy4 flash
BenchmarkDeepSeek V4 Pro 0813Hy4 flashLeader
LMArena WebDev1581Not measuredOnly DeepSeek V4 Pro 0813 measured
LiveCodeBench (Vals AI)87.5%Not measuredOnly DeepSeek V4 Pro 0813 measured
SWE-bench Verified (Vals AI)96.4%Not measuredOnly DeepSeek V4 Pro 0813 measured
Terminal-Bench 4 (Vals AI)14.1%15.2%Hy4 flash
Vibe Code Bench (Vals AI)82.3%74.8%DeepSeek V4 Pro 0813
Code migration (Vals AI)41.5%Not measuredOnly DeepSeek V4 Pro 0813 measured
ProgramBench (Vals AI)0.0%Not measuredOnly DeepSeek V4 Pro 0813 measured
IOI (Vals AI)51.6%Not measuredOnly DeepSeek V4 Pro 0813 measured
Terminal-Bench 2.1 (Vals AI)54.7%Not measuredOnly DeepSeek V4 Pro 0813 measured
Vibe Code Bench 1-100 (Vals AI)17.5%Not measuredOnly DeepSeek V4 Pro 0813 measured
SkillsBench (Vals AI)53.8%Not measuredOnly DeepSeek V4 Pro 0813 measured
CyberBench (Vals AI)72.9%Not measuredOnly DeepSeek V4 Pro 0813 measured
SciCode51.7%Not measuredOnly DeepSeek V4 Pro 0813 measured
Terminal-Bench 4.0 (Artificial Analysis run)14.1%Not measuredOnly DeepSeek V4 Pro 0813 measured
FrontierCode 2.017.1%Not measuredOnly DeepSeek V4 Pro 0813 measured
FrontierCode28.5%Not measuredOnly DeepSeek V4 Pro 0813 measured
WeirdML66.2%Not measuredOnly DeepSeek V4 Pro 0813 measured
MirrorCode33.3%Not measuredOnly DeepSeek V4 Pro 0813 measured
ALE-Bench1403.2Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: coding1467Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: software and IT1460Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena WebDev: React1585Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena WebDev: HTML1562Not measuredOnly DeepSeek V4 Pro 0813 measured
GameBench 253.250.5DeepSeek V4 Pro 0813
Terminal-Bench 4.06.7%Not measuredOnly DeepSeek V4 Pro 0813 measured
Neither model has a result on the other 27 boards here
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
Intelligence benchmarks for DeepSeek V4 Pro 0813 and Hy4 flash
BenchmarkDeepSeek V4 Pro 0813Hy4 flashLeader
AA Intelligence Index23.1Not measuredOnly DeepSeek V4 Pro 0813 measured
Epoch Capabilities Index156.0Not measuredOnly DeepSeek V4 Pro 0813 measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index30
Agentic work benchmarks for DeepSeek V4 Pro 0813 and Hy4 flash
BenchmarkDeepSeek V4 Pro 0813Hy4 flashLeader
SREBench (Vals AI)1.9%Not measuredOnly DeepSeek V4 Pro 0813 measured
Finance Agent (Vals AI)50.4%Not measuredOnly DeepSeek V4 Pro 0813 measured
Tax Agent Bench (Vals AI)26.8%23.3%DeepSeek V4 Pro 0813
Legal Research Bench (Vals AI)40.9%38.5%DeepSeek V4 Pro 0813
Harvey's Legal Agent Benchmark (Vals AI)10.8%5.4%DeepSeek V4 Pro 0813
Excel Modeling Benchmark (Vals AI)52.8%Not measuredOnly DeepSeek V4 Pro 0813 measured
Terminal-Bench-Science (Vals AI)4.3%Not measuredOnly DeepSeek V4 Pro 0813 measured
APEX-Agents47.3%Not measuredOnly DeepSeek V4 Pro 0813 measured
Toolathlon-Verified74.4%Not measuredOnly DeepSeek V4 Pro 0813 measured
Neither model has a result on the other 35 boards here
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index13 boards, none shared
Reasoning benchmarks for DeepSeek V4 Pro 0813 and Hy4 flash
BenchmarkDeepSeek V4 Pro 0813Hy4 flashLeader
GPQA Diamond92.4%Not measuredOnly DeepSeek V4 Pro 0813 measured
GPQA Diamond (Artificial Analysis run)92.8%Not measuredOnly DeepSeek V4 Pro 0813 measured
Humanity's Last Exam (Artificial Analysis run)41.5%Not measuredOnly DeepSeek V4 Pro 0813 measured
HLE-Diamond (Artificial Analysis run)20.5%Not measuredOnly DeepSeek V4 Pro 0813 measured
CritPt v2, hard subset (Artificial Analysis run)28.5%Not measuredOnly DeepSeek V4 Pro 0813 measured
GPQA Diamond (Epoch AI run)91.7%Not measuredOnly DeepSeek V4 Pro 0813 measured
ARC-AGI90.5%Not measuredOnly DeepSeek V4 Pro 0813 measured
ARC-AGI-261.3%Not measuredOnly DeepSeek V4 Pro 0813 measured
Chess puzzles46.0%Not measuredOnly DeepSeek V4 Pro 0813 measured
Mystery game puzzles43.0%Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: hard prompts1456Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: expert1455Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: hard prompts in English1463Not measuredOnly DeepSeek V4 Pro 0813 measured
Neither model has a result on the other 13 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
Math benchmarks for DeepSeek V4 Pro 0813 and Hy4 flash
BenchmarkDeepSeek V4 Pro 0813Hy4 flashLeader
OTIS Mock AIME98.6%Not measuredOnly DeepSeek V4 Pro 0813 measured
FrontierMath, tiers 1 to 364.6%Not measuredOnly DeepSeek V4 Pro 0813 measured
FrontierMath, tier 426.8%Not measuredOnly DeepSeek V4 Pro 0813 measured
ProofBench50.0%Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: math1432Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: mathematical work1437Not measuredOnly DeepSeek V4 Pro 0813 measured
Neither model has a result on the other 8 boards here
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index11 boards, none shared
Knowledge benchmarks for DeepSeek V4 Pro 0813 and Hy4 flash
BenchmarkDeepSeek V4 Pro 0813Hy4 flashLeader
MMLU Pro87.0%Not measuredOnly DeepSeek V4 Pro 0813 measured
LegalBench (Vals AI)82.4%Not measuredOnly DeepSeek V4 Pro 0813 measured
CorpFin (Vals AI)65.4%Not measuredOnly DeepSeek V4 Pro 0813 measured
TaxEval (Vals AI)73.1%Not measuredOnly DeepSeek V4 Pro 0813 measured
MedCode (Vals AI)42.5%Not measuredOnly DeepSeek V4 Pro 0813 measured
AA-Omniscience-6.4Not measuredOnly DeepSeek V4 Pro 0813 measured
SimpleQA Verified52.9%Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: science1457Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: medicine and healthcare1447Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: legal and government1462Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: business and finance1442Not measuredOnly DeepSeek V4 Pro 0813 measured
Neither model has a result on the other 10 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
Web design benchmarks for DeepSeek V4 Pro 0813 and Hy4 flash
BenchmarkDeepSeek V4 Pro 0813Hy4 flashLeader
Design Arena: websitesNot measured1227Only Hy4 flash measured
Design Arena: UI componentsNot measured1262Only Hy4 flash measured
Design Arena: game devNot measured1202Only Hy4 flash measured
Design Arena: 3DNot measured1209Only Hy4 flash measured
Design Arena: all code categoriesNot measured1225Only Hy4 flash measured
Neither model has a result on the other 4 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index10
General benchmarks for DeepSeek V4 Pro 0813 and Hy4 flash
BenchmarkDeepSeek V4 Pro 0813Hy4 flashLeader
Vals Index53.0%50.8%DeepSeek V4 Pro 0813
MedScribe (Vals AI)80.2%Not measuredOnly DeepSeek V4 Pro 0813 measured
Long context reasoning (AA-LCR)80.3%Not measuredOnly DeepSeek V4 Pro 0813 measured
GDPval-AA1456Not measuredOnly DeepSeek V4 Pro 0813 measured
DTBench93.9%Not measuredOnly DeepSeek V4 Pro 0813 measured
LMCA45.5%Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text1445Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: instruction following1448Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: creative writing1438Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: English1458Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: non-English1429Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: Chinese1478Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: French1440Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: Spanish1437Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: Russian1445Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: multi-turn1452Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: longer queries1457Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: writing and language1442Not measuredOnly DeepSeek V4 Pro 0813 measured
LMArena Text: entertainment and media1424Not measuredOnly DeepSeek V4 Pro 0813 measured
Neither model has a result on the other 30 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for DeepSeek V4 Pro 0813 and Hy4 flash
BenchmarkDeepSeek V4 Pro 0813Hy4 flashLeader
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

DeepSeek V4 Pro 0813$0.0033Fits in one request
Hy4 flash$0.0001Fits in one request

Hy4 flash costs less

Repository review

50K input + 3K output tokens

DeepSeek V4 Pro 0813$0.077Fits in one request
Hy4 flash$0.0022Fits in one request

Hy4 flash costs less

Agent loop

200K input + 10K output tokens

DeepSeek V4 Pro 0813$0.302Fits in one request
Hy4 flash$0.0084Fits in one request

Hy4 flash 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 DeepSeek V4 Pro 0813 and Hy4 flash
SpecificationDeepSeek V4 Pro 0813Tencent Hy4 flash
MakerDeepSeekTencent
WeightsOpen weightsOpen weights
ReleasedAugust 12, 2026November 16, 2026
Context window1.05M1.05M
Input $/Mtok$1.31$0.035
Output $/Mtok$3.96$0.1405
Coding index56.144.5
Independent measurements behind it52
Cost per solved task$0.10Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V4 Pro 0813 better than Hy4 flash for coding?

Pick DeepSeek V4 Pro 0813 for coding: it scores 56.1 on the Vibe Coding Index against 44.5 for Hy4 flash. Pick Hy4 flash for output tokens 28.2x cheaper ($0.1405 vs $3.96 per million) and work measured by Terminal-Bench 4 (Vals AI), where it leads.

Which is cheaper, DeepSeek V4 Pro 0813 or Hy4 flash?

DeepSeek V4 Pro 0813 costs $1.31 per million input tokens and $3.96 per million output tokens; Hy4 flash costs $0.035 and $0.1405. Cost per solved task is known for DeepSeek V4 Pro 0813 only.

Which has the bigger context window, DeepSeek V4 Pro 0813 or Hy4 flash?

Neither: both take 1.05M tokens.

How many benchmarks is the DeepSeek V4 Pro 0813 vs Hy4 flash comparison based on?

7 benchmarks have a result for both models as of November 22, 2026. DeepSeek V4 Pro 0813's coding index rests on 5 independent benchmarks and Hy4 flash's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/deepseek-v4-pro-0813-vs-tencent-hy4-flash" title="DeepSeek V4 Pro 0813 vs Hy4 flash, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/deepseek-v4-pro-0813-vs-tencent-hy4-flash">LLMBenchmark.ai, DeepSeek V4 Pro 0813 vs Hy4 flash</a></p>