DeepSeek
DeepSeek V4 Flash 0420
ReasoningOpen weights
- Intelligence
- 24.2
- Rank 188 of 762 measured models
- Coding
- 56.2
- Rank 100 of 299 measured models
- Output speed
- 45.9 t/s
- Rank 369 of 420 measured models
- Price/1M
- $0.142
- Rank 140 of 716 measured models
- Context window
- 1M
- Rank 88 of 1283 measured models
- Cost per task
- $0.0983
- Rank 45 of 214 measured models
DeepSeek V4 Flash 0420 (Max) scores 24.2 on the Artificial Analysis Intelligence Index, ranking 188th of 762 measured models. The closest models on this index are GPT-5.4 mini (24.1), Solar Mini 4 (24.1), and MiMo-V2-Omni (23.9). At $0.142 per million tokens (3:1 input/output blended price), it is cheaper than GPT-5.4 mini ($1.69) and Solar Mini 4 ($0.175). With an output speed of 45.9 t/s, it is slower than GPT-5.4 mini (223 t/s) and Solar Mini 4 (118 t/s).
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- Ling-3.0-flash-VL
InclusionAI · $0.111 · 124 t/s
24.6+0.4$0.111124 t/sCompare - GPT-5.4 mini
OpenAI · $1.69 · 223 t/s
24.1−0.1$1.69223 t/sCompare - Solar Mini 4
Upstage · $0.175 · 118 t/s
24.1−0.1$0.175118 t/sCompare - MiMo-V2-Omni
Xiaomi · — · —
23.9−0.3——Compare
