DeepSeek V3 (March 2024) has a larger context window than Qwen3 235B A22B (128K tokens vs 33K tokens).
| Spec | DeepSeek V3 (March 2024) | Qwen3 235B A22B |
|---|---|---|
| Input price | $0.14/M tokens | — |
| Output price | $0.28/M tokens | — |
| Context window | 128K tokens | 33K tokens |
| Released | Mar 2024 | Apr 2025 |
| AIME | 59.4% | Competitive with o3-mini |
Not enough votes to call it. On the specs, nothing separates them.
Qwen3 235B A22B takes Image Generation.
DeepSeek V3 (March 2024) uses 73.2x more hedging
DeepSeek V3 (March 2024) has a larger context window than Qwen3 235B A22B (128K tokens vs 33K tokens).
| Spec | DeepSeek V3 (March 2024) | Qwen3 235B A22B |
|---|---|---|
| Input price | $0.14/M tokens | — |
| Output price | $0.28/M tokens | — |
| Context window | 128K tokens | 33K tokens |
| Released | Mar 2024 | Apr 2025 |
| AIME | 59.4% | Competitive with o3-mini |
Not enough votes to call it. On the specs, nothing separates them.
Qwen3 235B A22B takes Image Generation.
DeepSeek V3 (March 2024) uses 73.2x more hedging