DeepSeek V4 Flash 0423 is 12.7× cheaper per million tokens (blended). DeepSeek V4 Flash 0423 has the larger context window (1,024,000 vs 256,000 tokens, 4.0×).
DeepSeek V4 Flash 0423 vs Step 3.7 Flash
Live catalog fields. Best-in-row is highlighted.
DeepSeek V4 Flash 0423open
DeepSeek · deepseek__deepseek-v4-flash
Step 3.7 Flashopen
Stepfun · stepfun__step-3.7-flash
Identity
| Field | deepseek__deepseek-v4-flash | stepfun__step-3.7-flash |
|---|---|---|
| Provider | DeepSeek | Stepfun |
| Slug | deepseek__deepseek-v4-flash | stepfun__step-3.7-flash |
| Status | live | live |
| Open weights | Yes | Yes |
| License | — | — |
| HuggingFace | deepseek-ai/DeepSeek-V4-Flash | stepfun-ai/Step-3.7-Flash |
Quality
| Field | deepseek__deepseek-v4-flash | stepfun__step-3.7-flash |
|---|---|---|
| Quality index (0–100) | 86.1 | — |
Cost
| Field | deepseek__deepseek-v4-flash | stepfun__step-3.7-flash |
|---|---|---|
| Input $ / 1M tokens | $0.04 | $0.20 |
| Cached input $ / 1M | $0.01 | $0.04 |
| Output $ / 1M tokens | $0.07 | $1.15 |
Context
| Field | deepseek__deepseek-v4-flash | stepfun__step-3.7-flash |
|---|---|---|
| Max context tokens | 1,024,000 | 256,000 |
| Max output tokens | 384,000 | 230,400 |
Modalities
| Field | deepseek__deepseek-v4-flash | stepfun__step-3.7-flash |
|---|---|---|
| Input | text | text, image, video |
| Output | text | text |