token.app › Compare › Claude Haiku 4.5 vs GPT-5.6 Luna

Claude Haiku 4.5 vs GPT-5.6 Luna

GPT-5.6 Luna leads on 3 of 3 shared benchmarks. GPT-5.6 Luna is 8.9× cheaper per token on a blended 3:1 basis. GPT-5.6 Luna accepts the larger context window (1.1M).

Head to head
Blended price uses a 3:1 input:output mix
Claude Haiku 4.5GPT-5.6 Luna
ProviderAnthropicOpenAI
Input $/1M$1.00$0.100
Output $/1M$5.00$0.600
Blended $/1M$2.00$0.225
Context window200K1.1M
Max output64K128K
ReleasedOct 2025Jul 2026
LicenceProprietaryProprietary
GPQA Diamond 71.2% 91.6%
SimpleQA Verified 5.9% 41.7%
OTIS Mock AIME 66.7% 98.3%
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 Claude Haiku 4.5 if…
  • No clear advantage on the data we hold.
Choose GPT-5.6 Luna if…
  • Costs less per token — $0.225 vs $2.00 blended
  • Higher GPQA Diamond (91.6% vs 71.2%)
  • Higher SimpleQA Verified (41.7% vs 5.9%)
  • Higher OTIS Mock AIME (98.3% vs 66.7%)
  • Larger context window (1.1M)
  • Newer model (released Jul 2026)
FAQ
Which is better, Claude Haiku 4.5 or GPT-5.6 Luna?
Across the 3 benchmarks both models have independently-run scores for, Claude Haiku 4.5 leads on 0 and GPT-5.6 Luna leads on 3. Benchmarks are one input — price and context window are on this page too.
Is Claude Haiku 4.5 cheaper than GPT-5.6 Luna?
GPT-5.6 Luna is cheaper: $0.225 versus $2.00 per million tokens blended at 3:1 input:output. Input alone: $1.00 vs $0.100. Output alone: $5.00 vs $0.600.
What context windows do Claude Haiku 4.5 and GPT-5.6 Luna have?
Claude Haiku 4.5: 200,000 tokens. GPT-5.6 Luna: 1,050,000 tokens.
Who makes Claude Haiku 4.5 and GPT-5.6 Luna?
Claude Haiku 4.5 is made by Anthropic. GPT-5.6 Luna is made by OpenAI.
Related comparisons