← Back to dashboard

DeepSeek

DeepSeek
· 📅 2026-04-03
DeepSeek's chat model. Outstanding reliability at the lowest cost, with moderate bearish tendency.
💰Lowest Cost$0.0022/run
🛡️Highest Validity100.0%
DeepSeek is the most conservative model — smallest average valuation gap (-5.1%). Moderate daily volatility (2.60%/day). Wins: Lowest Cost, Highest Validity. Perfect validity rate (100%). Cheapest model at $0.0022/run.

Model Positioning

Calibration score vs estimate consistency
Calibrated & ConsistentConsistent but UncalibratedCalibrated but VolatileLow ScoresCalibrationConsistencyGPTCLAUDEGEMINIDEEPSEEKGROK
Compare:GPTClaudeGeminiGrok

Key Metrics

-5.1%
Avg Gap
57/100
Calibration
43%
Correction Rate
33%
Positive Bias %
⚡ Responsive
Volatility
24
Days Tracked

Bias History

Model bias over time vs other models

Prediction Accuracy

Directional accuracy vs actual price changes
Directional Accuracy
47.3%
Mean Absolute Error
20.3%
Prediction Bias
-5.1%
Hit Rate (±5%)
8.7%
0%25%50%75%100%3.3.6.3.11.3.16.3.20.3.25.3.30.3.2.4.
GPT52.7%
Claude48.3%
Gemini47%
DeepSeek47.3%
Grok48.9%
N = 529 · 1 Day

Current Estimates

Largest Positive Gap
+38.3%
+36.8%
+19.1%
+14.2%
+13.3%
Largest Negative Gap
-50.8%
-28.4%
-26.1%
-25.1%
-20.7%

Assumption Profile

6.5%
σ = 6.25
Revenue CAGR 5Y
9.5%
σ = 1.33
WACC
2.13%
σ = 0.25
Terminal Growth
23.2%
EBIT Margin Target

Validity

100.0%
552 / 552 validity rate

Cost & Performance

Avg Latency: 22.7s
Cost per Run: $0.0022
Total Cost: $1.20
~5356 + 667 tokens

Calibration Detail

57
/100
Correction Rate: 42.9%
Uncapped Deviation:
Total Valuations: 552
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