Gemma 3 4B is 1.8× cheaper per million tokens (blended). DeepSeek V4 Flash 0423 has the larger context window (1,048,575 vs 131,072 tokens, 8.0×).
DeepSeek V4 Flash 0423 vs Gemma 3 4B
Live catalog fields. Best-in-row is highlighted.
DeepSeek V4 Flash 0423open
DeepSeek · deepseek__deepseek-v4-flash
Gemma 3 4Bopen
Google · google__gemma-3-4b-it
Identity
| Field | deepseek__deepseek-v4-flash | google__gemma-3-4b-it |
|---|---|---|
| Provider | DeepSeek | |
| Slug | deepseek__deepseek-v4-flash | google__gemma-3-4b-it |
| Status | live | live |
| Open weights | Yes | Yes |
| License | — | — |
| HuggingFace | deepseek-ai/DeepSeek-V4-Flash | google/gemma-3-4b-it |
Quality
| Field | deepseek__deepseek-v4-flash | google__gemma-3-4b-it |
|---|---|---|
| Quality index (0–100) | — | — |
Cost
| Field | deepseek__deepseek-v4-flash | google__gemma-3-4b-it |
|---|---|---|
| Input $ / 1M tokens | $0.09 | $0.05 |
| Cached input $ / 1M | $0.02 | — |
| Output $ / 1M tokens | $0.18 | $0.10 |
Context
| Field | deepseek__deepseek-v4-flash | google__gemma-3-4b-it |
|---|---|---|
| Max context tokens | 1,048,575 | 131,072 |
| Max output tokens | 943,717 | 16,384 |
Modalities
| Field | deepseek__deepseek-v4-flash | google__gemma-3-4b-it |
|---|---|---|
| Input | text | text, image |
| Output | text | text |