Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

GPT-5.4 vs GPT-6 Sol

45 SHARED BENCHMARKS

Across 45 shared benchmarks, GPT-5.4 scores higher on 9 and GPT-6 Sol on 35, with 1 level. The widest gap is AA-Omniscience · Non-hallucination, where GPT-6 Sol scores 39.9 against 8.3. GPT-6 Sol is the cheaper of the two on tracked API pricing ($2.00 against $2.50 per million input tokens).

OPENAIVSOPENAI45 SHARED9–35 HEAD-TO-HEADNEWEST SCORE ADDED

At a glance

MakerOpenAIOpenAI
Released
Price per 1M tokens input / output$2.50 / $15.00$2.00 / $10.00
Cost of 1M in + 1M out$17.50$12.00 1.5× less
Head-to-head of 45 shared benchmarks9 wins35 wins
Scores tracked independently verified199 41 ◆126 37 ◆

Release dates per Artificial Analysis. Prices: Artificial Analysis for GPT-5.4; Artificial Analysis for GPT-6 Sol. ◆ = independently verified score.

Where each leadsby capability area · benchmark wins

AreaGPT-5.4WINSGPT-6 Sol
Reasoning05GPT-6 Sol leads 5 of 5 · widest: CritPt 23.4 vs 30.9
Coding13GPT-6 Sol leads 3 of 4 · widest: LMArena · WebDev 1466 vs 1681
Agentic02GPT-6 Sol leads 2 of 2 · widest: AA IT-Bench SRE 34.5 vs 49.4
Factuality03GPT-6 Sol leads 3 of 3 · widest: AA-Omniscience · Non-hallucination 8.3 vs 39.9
Instruction Following10GPT-5.4 leads 1 of 1
Long Context01GPT-6 Sol leads 1 of 1
Math01GPT-6 Sol leads 1 of 1
Multimodal01GPT-6 Sol leads 1 of 1 · widest: MMMU-Pro 78.4 vs 83.3

Biggest gaps

GPT-5.4 pulls furthest ahead on

No ratified-area lead of 3 points or more.

GPT-6 Sol pulls furthest ahead on

  1. AA-Omniscience · Non-hallucination39.9 vs 8.3
  2. AA IT-Bench SRE49.4 vs 34.5
  3. GDPVal49.3 vs 36.6

Every shared benchmark45 · grouped by area

Reasoning 5

BenchmarkGPT-5.4MARGINGPT-6 Sol
ARC-AGI-30.2 ◆◆ 4.6
ARC-AGI-273.3◆ 89.6
CritPt23.430.9
LiveBench · Reasoning88.1 ◆◆ 88.7

Coding 4

BenchmarkGPT-5.4MARGINGPT-6 Sol
LMArena · WebDev1466 ◆◆ 1681
LiveBench · Coding77.5 ◆◆ 81.8
SciCode56.657.6

Agentic 2

BenchmarkGPT-5.4MARGINGPT-6 Sol
GDPVal36.649.3

Factuality 3

Instruction Following 1

Long Context 1

BenchmarkGPT-5.4MARGINGPT-6 Sol
AA-LCR8283.7

Math 1

BenchmarkGPT-5.4MARGINGPT-6 Sol
LiveBench · Mathematics94.2 ◆◆ 96.4

Multimodal 1

BenchmarkGPT-5.4MARGINGPT-6 Sol
MMMU-Pro78.483.3

Other shared benchmarks 27

BenchmarkGPT-5.4MARGINGPT-6 Sol
ARC-AGI-193.7 ◆◆ 95.5
Dynamic Benchmarks - Emotional reliance11
Dynamic Benchmarks - Mental health0.91
Dynamic Benchmarks - Self-harm11
Image input evaluations - extremism11
Image input evaluations - harms-erotic11
Image input evaluations - hate11
Image input evaluations - self-harm11
livebench_data_analysis79.3 ◆◆ 81.2
livebench_language82.6 ◆◆ 85.3
Production Benchmarks - Gore0.80.9
U18 evaluations - Age-restricted goods, services, and dangerous challenges / activities0.80.9
U18 evaluations - Eating Disorders0.70.9
U18 evaluations - Emotional Reliance0.90.9
U18 evaluations - Gore0.80.9
U18 evaluations - Self Harm11
U18 evaluations - Sexual Content0.91
vectara_answer_rate99.9 ◆◆ 100

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. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (direct, direct), otherwise the lowest tracked offer.

Compare GPT-5.4 withALL PAIRINGS →