| Model | Input / output per 1M | Log probabilities | Quantization |
|---|---|---|---|
| GLM 5.3 | $1.30 / $4.40 | yes | fp8 |
| gpt-oss-120b | $0.03 / $0.17 | yes | bf16 |
| gpt-oss-20b | $0.02 / $0.10 | yes | fp4 |
| Qwen3.8 27B | $0.25 / $2.20 | yes | fp8 |
| Llama 3.3 70B Instruct | $0.20 / $0.52 | yes | fp8 |
| Model | Input / output per 1M | Log probabilities | Quantization |
|---|---|---|---|
| GLM 5.3 | $1.30 / $4.40 | yes | fp8 |
| gpt-oss-120b | $0.03 / $0.17 | yes | bf16 |
| gpt-oss-20b | $0.02 / $0.10 | yes | fp4 |
| Qwen3.8 27B | $0.25 / $2.20 | yes | fp8 |
| Llama 3.3 70B Instruct | $0.20 / $0.52 | yes | fp8 |