Skip to content

GPT-6 Luna

OpenAIProprietaryReasoningVerifiedReleased Sep 22, 2026Input: file, image, text
Compare with GPT-5.2 ProSource scores ↗OpenRouter ↗
Overall score
66.6
Rank 32 of 222 · 62.5–70.7
Context window
1.05M
1,050,000 tokens
Price per 1M tokens
$0.10 in · $0.50 out
✓ Verified on OpenAI's pricing page
Blended price
$0.20
3:1 input:output tokens
Output speed
—
Not measured yet

Category scores

Scores per capability area with the model's rank among ranked models in that category.

  1. 1Math83.7
  2. 2Multimodal81.9
  3. 3Agentic & tool use#3873.3
  4. 4Knowledge#2768.2
  5. 5Coding#3866.3
  6. 6Reasoning#2459.2
Higher is better. #N is the model's rank in that category, where it is eligible to be ranked.

Benchmark results (21)

As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.

Knowledge

HealthBench Professional60.8#7 of 16
HealthBench Hard31.4#8 of 12
HealthBench (raw)50.0#6 of 7
HealthBench (length-adjusted)54.5#6 of 7
HealthBench Professional (raw)61.2#6 of 7

Agentic & tool use

MazeBench0#20 of 42
Hermes Index33.9#13 of 16
Hermes Bench70.5#11 of 16
ExploitGym11.6#11 of 14

Reasoning

ARC-AGI-259.3#30 of 41
ARC-AGI-186.67#27 of 36
ARC-AGI-30.1#21 of 22
ARC-AGI-3 Provider Adapter0.59#8 of 8

Multimodal

ParseBench65.77#20 of 37
ExtractBench (LlamaIndex)(lower is better)91.01#17 of 24
ExtractBench Pro0#5 of 5

Coding

DeepSWE66.6#28 of 50
Bug Hunt Bench18.3#20 of 23

Math

FrontierMath v2 (Tiers 1-3)78.947#16 of 64
FrontierMath v2 (Tier 4)56.098#16 of 51

Pricing by source

USD per 1M tokens.

SourceInputOutputCached in
✓ Official (OpenAI)
checked Oct 17, 2026 · gpt-6-luna
$0.10$0.50$0.010
BenchLM list price$0.10$0.50$0.010
OpenRouter openai/gpt-6-luna$0.10$0.50$0.010

Standard tier, prompts under 272K tokens.

OpenAI's GPT-6 Luna model page lists $0.10 input / $0.01 cached input / $0.125 cache writes / $0.50 output per million tokens. Prompts above 272K input tokens are charged at 2x input and cache rates and 1.5x output for the full request; regional processing adds 10% where available; Batch and Flex are 50% of Standard and Fast mode is 2x.

Arena ratings

Human-preference ratings from LMArena. Best-rated variant shown.

Text(max)1392#166 · 49.6K votes
WebDev (coding)(Max)1578#38 · 21.6K votes
Vision(Max)1216#85 · 3.7K votes
Search(max)1160#32 · 10.8K votes

Speed

Measured by us through OpenRouter (median of trials).

Not measured yet.

Related models

Same family

Similar score