Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

Seed 1.8 vs Seed2.1

17 SHARED BENCHMARKS

Across 17 shared benchmarks, Seed 1.8 scores higher on 1 and Seed2.1 on 16. The widest gap is ERQA, where Seed2.1 scores 71.3 against 58.8.

BYTEDANCEVSX17 SHARED1–16 HEAD-TO-HEADNEWEST SCORE ADDED

At a glance

MakerByteDancex
Released——
Price per 1M tokens input / output$0.25 / $2.00—
Cost of 1M in + 1M out$2.25—
Head-to-head of 17 shared benchmarks1 wins16 wins
Scores tracked independently verified40 0 ◆50 0 ◆

Prices: deepinfra for Seed 1.8. ◆ = independently verified score.

Where each leadsby capability area · benchmark wins

AreaSeed 1.8WINSSeed2.1
Multimodal03Seed2.1 leads 3 of 3 · widest: MMMU-Pro 73.2 vs 80.1

Biggest gaps

Seed 1.8 pulls furthest ahead on

No ratified-area lead of 3 points or more.

Seed2.1 pulls furthest ahead on

  1. MMMU-Pro80.1 vs 73.2
  2. BLINK79.4 vs 74.3

Every shared benchmark17 · grouped by area

Multimodal 3

BenchmarkSeed 1.8MARGINSeed2.1
BLINK74.379.4
MathVista87.790.5
MMMU-Pro73.280.1

Other shared benchmarks 14

BenchmarkSeed 1.8MARGINSeed2.1
DUDE69.483.1
DynaMath61.568.1
ERQA58.871.3
LVBench7376.8
Minerva62.465.9
MotionBench70.674.8
OVBench65.169.7
SimpleVQA65.471.1
SuperGPQA64.867.4
TOMATO60.856.8
TVBench71.577.2

Each score is the one the model's own page shows — the most authoritative tracked result for that benchmark, on the benchmark's standard methodology (source-attributed). ↓ marks lower-is-better metrics; ◆ an independently verified score. Open either model for its full surface, provenance and pricing.

Compare Seed2.1 withALL PAIRINGS →