gpt-oss-20b is 3.8× cheaper per million tokens (blended). GPT Luna Latest has the larger context window (1,050,000 vs 131,072 tokens, 8.0×). gpt-oss-20b is open-weights; the other is not.
gpt-oss-20b vs GPT Luna Latest
Live catalog fields. Best-in-row is highlighted.
gpt-oss-20bopen
OpenAI · openai__gpt-oss-20b
GPT Luna Latest
~openai · ~openai__gpt-luna-latest
Identity
| Field | openai__gpt-oss-20b | ~openai__gpt-luna-latest |
|---|---|---|
| Provider | OpenAI | ~openai |
| Slug | openai__gpt-oss-20b | ~openai__gpt-luna-latest |
| Status | live | live |
| Open weights | Yes | No |
| License | — | — |
| HuggingFace | openai/gpt-oss-20b | — |
Quality
| Field | openai__gpt-oss-20b | ~openai__gpt-luna-latest |
|---|---|---|
| Quality index (0–100) | — | — |
Cost
| Field | openai__gpt-oss-20b | ~openai__gpt-luna-latest |
|---|---|---|
| Input $ / 1M tokens | $0.03 | $0.10 |
| Cached input $ / 1M | $0.03 | $0.01 |
| Output $ / 1M tokens | $0.13 | $0.50 |
Context
| Field | openai__gpt-oss-20b | ~openai__gpt-luna-latest |
|---|---|---|
| Max context tokens | 131,072 | 1,050,000 |
| Max output tokens | 117,964 | 128,000 |
Modalities
| Field | openai__gpt-oss-20b | ~openai__gpt-luna-latest |
|---|---|---|
| Input | text | file, image, text |
| Output | text | text |