Noometry

Model comparison

DeepSeek V4 Flash vs ERNIE 5.1

DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 43.8 on the Noometry Index.

Last verified . 18 shared benchmarks.

DeepSeek V4 Flash DeepSeek

53.6

Rank #35 Confirmed

ERNIE 5.1 Baidu

43.8

Rank #83 Confirmed

Summary

  • They share 18 benchmarks with published results for both. DeepSeek V4 Flash scores higher in 4 categories and ERNIE 5.1 in 4 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where DeepSeek V4 Flash leads 53.7 to 21.9.
  • The biggest single-benchmark swing is NYT Connections (extended): 89.6% for DeepSeek V4 Flash and 23.4% for ERNIE 5.1.
  • DeepSeek V4 Flash has downloadable open weights; the other is API-only.

Side by side

DeepSeek V4 FlashERNIE 5.1
ProviderDeepSeekBaidu
Noometry Index53.643.8
Released2026-04-24—
WeightsOpenProprietary
Context window1M—
Max output393K—
Input $ / M tokens$0.15—
Output $ / M tokens$0.60—
Results tracked4119

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding DeepSeek V4 Flash leads

DeepSeek V4 Flash: 47.9 (#59), ERNIE 5.1: 44.0 (#76)

BenchmarkDeepSeek V4 FlashERNIE 5.1
LMArena Coding14571488
FrontierCode18.8%—
LMArena WebDev1582—
SciCode49.9%—
WeirdML63%—
ALE-Bench1,306—

Agentic & Tool Use Not comparable

DeepSeek V4 Flash: —, ERNIE 5.1: —

BenchmarkDeepSeek V4 FlashERNIE 5.1
LMArena Search—1227

Reasoning DeepSeek V4 Flash leads

DeepSeek V4 Flash: 53.7 (#30), ERNIE 5.1: 21.9 (#211)

BenchmarkDeepSeek V4 FlashERNIE 5.1
NYT Connections (extended)89.6%23.4%
LMArena Hard Prompts14441481
ARC-AGI-261.4%—
SimpleBench61.1%—
Kagi LLM Benchmark52.2%—
ARC-AGI-189%—
CritPt16.6%—
Chess Puzzles33%—
Mystery Game Puzzles34%—
DTBench90.9%—
LMCA41.7%—
Epoch Capabilities Index154.49—

Math DeepSeek V4 Flash leads

DeepSeek V4 Flash: 60.3 (#37), ERNIE 5.1: 40.3 (#92)

BenchmarkDeepSeek V4 FlashERNIE 5.1
LMArena Math14271481
FrontierMath (Tiers 1-3)57.5%—
FrontierMath Tier 424.4%—
MathArena Final-Answer Competitions76.5%—
OTIS Mock AIME 2024-202594.4%—
ProofBench56%—

Knowledge DeepSeek V4 Flash leads

DeepSeek V4 Flash: 55.4 (#48), ERNIE 5.1: 41.9 (#102)

BenchmarkDeepSeek V4 FlashERNIE 5.1
LMArena Expert14411492
GPQA Diamond91%—
SimpleQA Verified33.6%—

Multilingual ERNIE 5.1 leads

DeepSeek V4 Flash: 53.0 (#72), ERNIE 5.1: 55.5 (#29)

BenchmarkDeepSeek V4 FlashERNIE 5.1
LMArena Non-English14201454
LMArena Chinese14681508
LMArena French14391488
LMArena German14181470
LMArena Japanese14061422
LMArena Korean13841427
LMArena Russian14281459
LMArena Spanish14361473

Instruction Following ERNIE 5.1 leads

DeepSeek V4 Flash: 74.9 (#81), ERNIE 5.1: 76.7 (#37)

BenchmarkDeepSeek V4 FlashERNIE 5.1
LMArena Instruction Following14211460

Long Context Too close to call

DeepSeek V4 Flash: 43.8 (#85), ERNIE 5.1: 44.7 (#59)

BenchmarkDeepSeek V4 FlashERNIE 5.1
LMArena Longer Query14341462

Writing & Preference ERNIE 5.1 leads

DeepSeek V4 Flash: 63.8 (#61), ERNIE 5.1: 65.1 (#52)

BenchmarkDeepSeek V4 FlashERNIE 5.1
LMArena Text14321468
LMArena Creative Writing14031441
LMArena Multi-Turn14491471
EQ-Bench Creative Writing1559—

Frequently asked questions

Is DeepSeek V4 Flash better than ERNIE 5.1?

DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 43.8 on the Noometry Index.

Is DeepSeek V4 Flash or ERNIE 5.1 better for coding?

DeepSeek V4 Flash scores higher on coding benchmarks: 47.9 versus 44.0 in the Noometry coding category.

How many benchmarks do DeepSeek V4 Flash and ERNIE 5.1 share?

18 benchmarks have published results for both models. DeepSeek V4 Flash has 41 scored results on Noometry and ERNIE 5.1 has 19.

Related comparisons

Go deeper