DeepSeek V4.1 Flash
DeepSeek V4.1 Flash supports text and image input, reasoning, tool calling, and structured output with a 1M-token context window. This is a rate-limited beta with limited capacity, intended for testing rather than production use.
- Reasoning
- Vision
- Tool Calling
- Structured Output
Added Sep 8, 2026
Model weightsPricing
Auto routing · per 1M tokens- Input
- $0.022
- Output
- $0.40
- Cache read
- $0.0040
Specifications
- Context window
- 1M
- Max output
- 384K
- Avg output (7d)
- 530 tokens
- Longer than 46% of models
Benchmarks
Benchmarks
Sourced from Vectara.
Hallucination Rate
4.2%
Better than 95% of models compared
Factual Consistency
95.8%
Better than 95% of models compared
Answer Rate
99.9%
Better than 76% of models compared
Avg Summary Length
Average generated summary length
108.2
Last updated 2026-10-22 · Matched as deepseek-ai/DeepSeek-V4.1-Flash
Vectara LeaderboardProviders
Choose or restrict providers with the API. Provider-selection prices include the selection markup. Provider selection guide · Pricing API (JSON, USD per 1K tokens)
 | $0.022 | $0.40 | $0.0040 | N/A | N/A |
| $0.022 | $0.40 | $0.0040 | N/A | N/A |
| $0.0063 | $0.525 | $0.00525 | 0.5s | 97 tps | |
| $0.0231 | $0.42 | $0.0042 | 0.6s | 86 tps | |
| $0.0305 | $0.121 | $0.0105 | N/A | N/A | |
| $0.0315 | $0.126 | $0.0063 | N/A | N/A | |
| $0.315$0.0532 | $1.26$0.213 | $0.0063$0.0011 | 1.1s | 90 tps | |
| $0.315$0.0787 | $1.26$0.315 | $0.0315$0.00788 | 0.6s | 117 tps | |
| $0.136 | $0.682 | $0.0021 | 0.5s | 131 tps | |
| $0.315$0.142 | $1.26$0.567 | $0.0315$0.0142 | 0.7s | 93 tps | |
| $0.157$0.142 | $0.63$0.567 | $0.0105$0.00945 | N/A | N/A | |
| $0.147 | $0.441 | $0.0044 | 0.5s | 111 tps | |
| $0.148 | $0.592 | $0.0030 | N/A | N/A | |
| $0.229$0.154 | $0.918$0.617 | $0.0046$0.0031 | 1.8s | 64 tps | |
| $0.157 | $0.525 | $0.157 | 0.9s | 65 tps | |
| $0.157 | $0.63 | $0.0158 | N/A | N/A | |
| $0.157 | $0.63 | $0.0105 | N/A | N/A | |
| $0.164 | $1.26 | $0.00504 | 0.7s | 168 tps | |
| $0.315$0.189 | $1.26$0.756 | $0.0063$0.0038 | 2.4s | 55 tps | |
| $0.20 | $0.682 | $0.0126 | 0.2s | 199 tps | |
| $0.315$0.205 | $1.26$0.819 | $0.0063$0.0041 | 1s | 104 tps | |
| $0.219 | $0.90 | N/A | N/A | N/A | |
| $0.252$0.221 | $1.01$0.882 | $0.00504$0.0044 | N/A | N/A | |
| $0.362$0.221 | $1.45$0.882 | $0.00724$0.0044 | 1.1s | 125 tps | |
| $0.289 | $1.26 | $0.0063 | N/A | N/A | |
| $0.289 | $1.26 | $0.0063 | N/A | N/A | |
| $0.305 | $1.26 | $0.00735 | 0.5s | 189 tps | |
| $0.315 | $1.26 | $0.00735 | 0.2s | 166 tps | |
| $0.315 | $1.26 | $0.0063 | 0.5s | 155 tps | |
| $0.315 | $1.26 | $0.0315 | 0.9s | 120 tps | |
| $0.315 | $1.26 | $0.00945 | N/A | N/A | |
| $0.315 | $1.26 | $0.0315 | 0.4s | 146 tps | |
| $0.315 | $1.26 | $0.0105 | 0.9s | 142 tps | |
| $0.315 | $1.26 | $0.0525 | N/A | N/A | |
| $0.315 | $1.26 | $0.0063 | N/A | N/A | |
| $0.315 | $1.26 | $0.0063 | 0.6s | 168 tps | |
| $0.315 | $1.26 | $0.0063 | 1.6s | 82 tps | |
| $0.315 | $1.26 | $0.0063 | 1.7s | 160 tps | |
| $0.315 | $1.26 | $0.00788 | 0.8s | 159 tps |
Privacy:ZDR, prompts not retainedNo training on promptsRetention and training unclearBest latency and throughput are highlighted.
Batch API
Batch rows for this model use normal routing and cost the same as standard API requests. Set provider in a row to choose a provider.