gpt-oss-20b is 2.3× cheaper per million tokens (blended). Llama Guard 4 12B has the larger context window (163,840 vs 131,072 tokens).
Llama Guard 4 12B vs gpt-oss-20b
Live catalog fields. Best-in-row is highlighted.
Llama Guard 4 12Bopen
Meta · meta-llama__llama-guard-4-12b
gpt-oss-20bopen
OpenAI · openai__gpt-oss-20b
Identity
| Field | meta-llama__llama-guard-4-12b | openai__gpt-oss-20b |
|---|---|---|
| Provider | Meta | OpenAI |
| Slug | meta-llama__llama-guard-4-12b | openai__gpt-oss-20b |
| Status | live | live |
| Open weights | Yes | Yes |
| License | — | — |
| HuggingFace | meta-llama/Llama-Guard-4-12B | openai/gpt-oss-20b |
Quality
| Field | meta-llama__llama-guard-4-12b | openai__gpt-oss-20b |
|---|---|---|
| Quality index (0–100) | — | — |
Cost
| Field | meta-llama__llama-guard-4-12b | openai__gpt-oss-20b |
|---|---|---|
| Input $ / 1M tokens | $0.18 | $0.03 |
| Cached input $ / 1M | — | $0.03 |
| Output $ / 1M tokens | $0.18 | $0.13 |
Context
| Field | meta-llama__llama-guard-4-12b | openai__gpt-oss-20b |
|---|---|---|
| Max context tokens | 163,840 | 131,072 |
| Max output tokens | 16,384 | 117,964 |
Modalities
| Field | meta-llama__llama-guard-4-12b | openai__gpt-oss-20b |
|---|---|---|
| Input | image, text | text |
| Output | text | text |