Compare
2–4 model comparison. Best-in-row is highlighted.
DeepSeek V4 Pro 0423open
DeepSeek · deepseek__deepseek-v4-pro
Qwen3.8 27Bopen
Qwen · qwen__qwen3.8-27b
Identity
| Field | deepseek__deepseek-v4-pro | qwen__qwen3.8-27b |
|---|---|---|
| Provider | DeepSeek | Qwen |
| Slug | deepseek__deepseek-v4-pro | qwen__qwen3.8-27b |
| Status | live | live |
| Open weights | Yes | Yes |
| License | — | — |
| HuggingFace | deepseek-ai/DeepSeek-V4-Pro | Qwen/Qwen3.8-27B |
Quality
| Field | deepseek__deepseek-v4-pro | qwen__qwen3.8-27b |
|---|---|---|
| Quality index (0–100) | 88.2 | 89.0 |
Cost
| Field | deepseek__deepseek-v4-pro | qwen__qwen3.8-27b |
|---|---|---|
| Input $ / 1M tokens | $0.72 | $0.42 |
| Cached input $ / 1M | $0.06 | $0.09 |
| Output $ / 1M tokens | $1.44 | $3.00 |
Context
| Field | deepseek__deepseek-v4-pro | qwen__qwen3.8-27b |
|---|---|---|
| Max context tokens | 1,024,000 | 1,000,000 |
| Max output tokens | 384,000 | 131,072 |
Modalities
| Field | deepseek__deepseek-v4-pro | qwen__qwen3.8-27b |
|---|---|---|
| Input | text | text, image, video |
| Output | text | text |