token.app โ€บ Compare โ€บ GPT-4 Turbo vs Kimi K3

GPT-4 Turbo vs Kimi K3

Kimi K3 leads on 2 of 2 shared benchmarks. Kimi K3 is 2.5ร— cheaper per token on a blended 3:1 basis. Kimi K3 accepts the larger context window (1.0M).

Head to head
Blended price uses a 3:1 input:output mix
GPT-4 TurboKimi K3
ProviderOpenAIMoonshot AI
Input $/1M$10.00$3.00
Output $/1M$30.00$15.00
Blended $/1M$15.00$6.00
Context window128K1.0M
Max output4Kโ€”
ReleasedApr 2024Jul 2026
LicenceProprietaryOpen weights
GPQA Diamond 46.6% 93.1%
OTIS Mock AIME 6.7% 97.2%
Benchmark scores independently run and published by Epoch AI, 'AI Benchmarking Hub', used under CC BY 4.0. Only benchmarks with a score for BOTH models are compared.
Choose GPT-4 Turbo ifโ€ฆ
  • No clear advantage on the data we hold.
Choose Kimi K3 ifโ€ฆ
  • Costs less per token โ€” $6.00 vs $15.00 blended
  • Higher GPQA Diamond (93.1% vs 46.6%)
  • Higher OTIS Mock AIME (97.2% vs 6.7%)
  • Larger context window (1.0M)
  • Open weights โ€” self-hostable
  • Newer model (released Jul 2026)
FAQ
Which is better, GPT-4 Turbo or Kimi K3?
Across the 2 benchmarks both models have independently-run scores for, GPT-4 Turbo leads on 0 and Kimi K3 leads on 2. Benchmarks are one input โ€” price and context window are on this page too.
Is GPT-4 Turbo cheaper than Kimi K3?
Kimi K3 is cheaper: $6.00 versus $15.00 per million tokens blended at 3:1 input:output. Input alone: $10.00 vs $3.00. Output alone: $30.00 vs $15.00.
What context windows do GPT-4 Turbo and Kimi K3 have?
GPT-4 Turbo: 128,000 tokens. Kimi K3: 1,048,576 tokens.
Who makes GPT-4 Turbo and Kimi K3?
GPT-4 Turbo is made by OpenAI. Kimi K3 is made by Moonshot AI.
Related comparisons