readable AI benchmarks (simplified)
OpenAI
GPT-5.6 Luna (Max)
GPT-5.6 Luna (Max) has a Quality Score of 51.1, ranking 98th among 571 scored models. Per 1M tokens, pricing is $0.200 input, $1.20 output, $0.020 cached input, and $0.250 cache write. It accepts text, image, and pdf input, outputs text, and has a 1M token context window. Its Reliability Score is 44.9, ranking 186th among 573 scored models, above average (total average: 36.5). Its Value Score is 43.7, high compared with the total Value average of 31.4.
Model specification
- Reasoning
- reasoning
- Input modalities
- text, image, pdf
- Output modalities
- text
- Context window
- 1M tokens
- Weights
- closed weights
Token prices USD per 1M tokens
- Input
- $0.200
- Output
- $1.20
- Cached Input
- $0.020
- Cache write
- $0.250
- Cost per task
- $0.178
Capability
- Intelligence
- 37.3
- Coding
- not available
- Agentic
- not available
- Omniscience
- -10.3
- Correct
- 42.7%
- Blended price
- $0.174
Answer outcomes
Fully graded outcomes (Correct + Incorrect): 95.8%
Artificial Analysis benchmarks
- GDPval-AA v2
- 48%
- τ³-Banking
- 31%
- SciCode
- 54%
- Humanity’s Last Exam
- 39%
- GPQA Diamond
- 91%
- CritPt
- 21%
- AA-Omniscience
- 45%
- AA-LCR
- 84%
DeepSWE operational cross-check
- Pass@1
- 67.2%
- Avg steps/scored attempt
- 101.68
- Mean cost/scored attempt (USD)
- $3.03
- Mean output tokens/scored attempt
- 73,399.708
- Mean duration/scored attempt (s)
- 1123 s
Mean cost/scored attempt is the official feed's mean_cost_usd: mean benchmark cost in USD per scored attempt under the source's cost basis. It is not Artificial Analysis cost per task and not token price (USD per 1M tokens). Mean output tokens/scored attempt is mean_output_tokens, also averaged over every scored attempt including scored failures. Mean duration/scored attempt is mean_duration_seconds; it is informational only, and wall-clock varies with provider and host load. These three fields do not affect Quality Score, Value Score, or Reliability. Epoch General ECI's benchmark basket includes DeepSWE v1.1, so Epoch General ECI and this DeepSWE cross-check are not independent evidence.
Similar models
- GPT-6 Luna (Xhigh)OpenAIQuality Score 51.1Value Score 46.7
- GPT-5.6 Terra (High)OpenAIQuality Score 50.5Value Score 33.0
- GPT-5.6 Terra (Xhigh)OpenAIQuality Score 53.4Value Score 35.0
- GPT-6 Luna (Max)OpenAIQuality Score 54.5Value Score 49.8
- GPT-5.6 Luna (Xhigh)OpenAIQuality Score 48.9Value Score 41.8
- GPT-5.3 Codex (Xhigh)OpenAIQuality Score 52.7Value Score 34.2