readable AI benchmarks (simplified)
OpenAI
GPT-5.6 Luna (Low)
GPT-5.6 Luna (Low) has a Quality Score of 37.5, ranking 213th among 572 scored models. Per 1M tokens, pricing is $0.200 input, $1.20 output, $0.020 cached input, and $0.250 cache write. It has a 1M token context window. Its Reliability Score is 42.7, ranking 214th among 574 scored models, above average (total average: 36.5). Its Value Score is 32.0, around average compared with the total Value average of 31.5.
Model specification
- Reasoning
- reasoning
- Input modalities
- none
- Output modalities
- none
- 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
- $0.0098
Capability
- Intelligence
- 21.0
- Coding
- not available
- Agentic
- not available
- Omniscience
- -14.7
- Correct
- 39.6%
- Blended price
- $0.174
Answer outcomes
Fully graded outcomes (Correct + Incorrect): 93.9%
Artificial Analysis benchmarks
- GDPval-AA v2
- 25%
- τ³-Banking
- 13%
- SciCode
- 46%
- Humanity’s Last Exam
- 20%
- GPQA Diamond
- 84%
- CritPt
- 3%
- AA-Omniscience
- 43%
- AA-LCR
- 70%
DeepSWE operational cross-check
- Pass@1
- 1.5%
- Avg steps/scored attempt
- 12.46
- Mean cost/scored attempt (USD)
- $0.072
- Mean output tokens/scored attempt
- 3,127.754
- Mean duration/scored attempt (s)
- 77 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-5 mini (Medium)OpenAIQuality Score 38.1Value Score 31.1
- GPT-5 (Low)OpenAIQuality Score 38.3Value Score 25.9
- GPT-5.1 Codex mini (High)OpenAIQuality Score 36.6Value Score 29.9
- o3OpenAIQuality Score 36.7Value Score 24.3
- GPT-6 Luna (Low)OpenAIQuality Score 39.3Value Score 35.9
- GPT-5.4 nano (Medium)OpenAIQuality Score 35.9Value Score 30.6