DeepSeek
DeepSeek V4.1 Flash
Open weights
- Intelligence
- 24.7
- Rank 192 of 759 measured models
- Output speed
- 248 t/s
- Rank 28 of 418 measured models
- Price/1M
- $0.525
- Rank 298 of 694 measured models
- Context window
- 1M
- Rank 92 of 1272 measured models
- Cost per task
- $0.234
- Rank 90 of 211 measured models
DeepSeek V4.1 Flash (Non-reasoning) scores 24.7 on the Artificial Analysis Intelligence Index, ranking 192nd of 759 measured models. The closest models on this index are Claude Sonnet 4.6 (24.7), GPT-5.1 (24.7), and Solar Open2 250B (24.7). At $0.525 per million tokens (3:1 input/output blended price), it is cheaper than Claude Sonnet 4.6 ($6) and GPT-5.1 ($3.44). With an output speed of 248 t/s, it is faster than Claude Sonnet 4.6 (48.7 t/s) and GPT-5.1 (138 t/s).
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- Claude Sonnet 4.6
Anthropic · $6 · 48.7 t/s
24.7$648.7 t/sCompare - GPT-5.1
OpenAI · $3.44 · 138 t/s
24.7$3.44138 t/sCompare - Solar Open2 250B
Upstage · — · —
24.7——Compare - Ling-3.0-flash-VL
InclusionAI · $0.111 · 124 t/s
24.6−0.1$0.111124 t/sCompare
