DeepSeek V4.1 Flash TEE
DeepSeek V4.1 Flash supports text and image input, reasoning, tool calling, and structured output with a 1M-token context window. This route runs through Tinfoil attested inference inside a Trusted Execution Environment.
- Reasoning
- Vision
- Tool Calling
- Parallel Tool Calls
- Structured Output
Added Sep 12, 2026
Model weightsPricing
Auto routing · per 1M tokens- Input
- $0.65
- Output
- $1.45
- Cache read
- $0.13
Specifications
- Context window
- 1M
- Max output
- 384K
- Avg output (7d)
- 498 tokens
- Longer than 44% of models
Benchmarks
Benchmarks
Sourced from Design Arena.
Best Category
3D
Best Elo
1357
Best Rank Percentile
Top 8%
Ranking Distribution
46,491 battles
60%
45%
30%
15%
0%
14.6%
First
39.1%
Second
35.5%
Third
10.8%
Fourth
Category Performance
Models Arena
3D
1357
Elo
58.2%
Win
150.3s
Avg Time
#13
Rank
Game Development
1334
Elo
54.5%
Win
154.8s
Avg Time
#14
Rank
UI Component
1318
Elo
52.9%
Win
111.2s
Avg Time
#20
Rank
Website
1311
Elo
50.6%
Win
137.2s
Avg Time
#16
Rank
Data Visualization
1294
Elo
50.4%
Win
111.2s
Avg Time
#15
Rank
ASCII Art
1251
Elo
50.9%
Win
112.2s
Avg Time
#22
Rank
SVG
1243
Elo
48.9%
Win
76.7s
Avg Time
#26
Rank
Updated Oct 14, 2026, 2:38 PM · Matched as Deepseek V4 1 Flash
Design ArenaProviders
Choose explicit providers for this model. Auto routing remains available as the default option.
Loading provider options…