DeepSeek
DeepSeek
· 2026-07-10
DeepSeek's chat model. Outstanding reliability at the lowest cost, with moderate bearish tendency.
Calibration and accuracy describe observed behavior, not predictive ability. Experimental research — not investment advice. Terms of Use
$Lowest Cost$0.0023/run
◆Highest Validity99.7%
DeepSeek is a moderate model with an average valuation gap of -16.0%. Moderate daily volatility (1.50%/day). Wins: Lowest Cost, Highest Validity. High validity rate (99.7%). Cheapest model at $0.0023/run.
Model Positioning
Calibration score vs estimate consistency
Key Metrics
-16%
Avg Gap
66/100
Calibration
20%
Correction Rate
23%
Positive Bias %
Responsive
Volatility
23
Days Tracked
Bias History
Model bias over time vs other models
Prediction Accuracy
Directional accuracy vs actual price changes
Directional Accuracy
45.8%
Mean Absolute Error
30.1%
Prediction Bias
-15.8%
Hit Rate (±5%)
17.5%
N = 504 · 1 Day
Current Estimates
Assumption Profile
6.5%
σ = 5.68
Revenue CAGR 5Y
9.6%
σ = 1.22
WACC
2.03%
σ = 0.14
Terminal Growth
23.6%
EBIT Margin Target
Validity
99.7%
2161 / 2168 validity rate
Cost & Performance
Avg Latency: 12.8s
Cost per Run: $0.0023
Total Cost: $4.88
~5628 + 668 tokens
Calibration Detail
66
/100
Correction Rate: 19.7%
Uncapped Deviation: —
Total Valuations: 528
Want these insights weekly?
Subscribe to AI Signals →AI Investor Barometer · DeepSeek← Back to dashboard
All content is generated by AI models and may contain errors. This is an experimental tool — not investment advice, research, or recommendation. Terms of Use · Privacy Policy