Market structure & setup evidence
Review NIFTY, BANKNIFTY and SENSEX with EMA alignment, VWAP, market structure and directional volume. Completed-candle evidence helps explain the setup.
Read the structure. Plan your risk. Check your mindset.
Use your own account-based PAPER journal, risk planner and reports. Connect Angel One for broker quotes and completed candles. The dashboard below remains a sample demo.
Observe the setup, plan the trade, then learn from the outcome. TradeDrishti brings market evidence, PAPER trade research and local AI review into one desktop workspace.
This browser workspace is an interactive demo. Download the desktop app for your own research and saved outcomes.
PAPER learning · ExperimentalSolid line: sample price · Dashed line: illustrative reference
In this bullish example, wait for a pullback and confirmation rather than chasing an extended move.
Quantity is before exchange lot sizing, fees and slippage. This preview does not place trades.
How are you approaching your next decision?
Choose a mindset to see a reflection prompt.
All market values and session metrics are sample data. No live feed, broker integration or AI inference is connected.
TradeDrishti connects the evidence before entry with the outcome after exit. These capabilities belong to the desktop application; the browser demo uses sample data.
Review NIFTY, BANKNIFTY and SENSEX with EMA alignment, VWAP, market structure and directional volume. Completed-candle evidence helps explain the setup.
Bullish pullback-reclaims become CE setup evidence; bearish rise-rejections become PE-buy evidence. Existing data, liquidity and risk checks still apply.
Review stops, targets, overlapping positions, cooldowns and loss limits. A promising setup still needs to pass the application's execution checks.
Keep captured trades, research observations and rejected setups distinct. Review entry evidence, exit reasons and favourable or adverse price movement.
Optional local text and chart-image models help review candle evidence and saved outcomes. The standard installer needs a separate AI runtime and model setup.
A local decision model refits on eligible closed PAPER outcomes and records TAKE, SKIP or WAIT with its version and reasons. It operates in shadow mode and does not control orders.
The learner's current target is gross PAPER P&L. Shadow predictions and a small historical sample do not establish live performance or profit after charges.
The desktop workflow connects market evidence, risk checks and saved outcomes. Practise in PAPER mode and review why a trade was taken or skipped.
Start with direction, structure and the current regime.
Buy on dips in a bullish regime. Consider puts on rises in a bearish regime. Be selective in sideways markets.
Review optional local AI commentary alongside the original candle evidence, risk reasons and uncertainty.
Compare entry reasoning with the closed PAPER result. Keep rejected research and forced samples separate from normal training evidence.
Download directly from our public GitHub Releases. No GitHub login is needed for these public downloads. Choose the installer for a normal setup, or the portable ZIP for an extracted app.
Windows 10 / 11 · 64-bit
72.4 MB · Python bundled
The filename retains the previous TPS application name. The application is TradeDrishti 1.6.6.
Windows 10 / 11 · 64-bit
105.1 MB · Extract before running
Extract the entire archive to a folder, then run the application inside. Do not run it from inside the ZIP.
Live market data needs internet and your own broker access. No broker credentials or personal trading records are supplied.
The browser demo cannot place orders. The desktop learner's shadow decisions do not enable automatic live execution.
The standard EXE and ZIP do not include the multi-GB Ollama runtime or models. For local text and chart-image review, install Ollama for Windows, then download the configured models: qwen3:1.7b and qwen3-vl:2b. Keep Ollama running on this PC at its default local address (127.0.0.1:11434). In TradeDrishti, open Reports & Learning → Local AI / Chart Vision and verify the runtime and model status. Initial model downloads need internet; inference runs locally once setup is complete.
After installing Ollama, open PowerShell and run:
ollama pull qwen3:1.7b
ollama pull qwen3-vl:2b
ollama listConfirm both models appear in the list. Keep the Ollama application open; if its local service is stopped, start it with ollama serve.
PAPER Learning itself does not need Ollama. Model speed depends on your PC; machines with limited RAM may run one model at a time. The 3.36 GB full offline AI installer is not currently hosted here.
No. Choose the named Windows EXE or Portable ZIP in the release assets. GitHub's automatic source archives are repository snapshots, not the ready-to-run Windows application.
These downloads are for Windows 10/11 64-bit. No macOS, Linux or mobile installer is offered on this page.
Compare the file's SHA-256 hash with the linked checksum file. On Windows PowerShell, use Get-FileHash -Algorithm SHA256 -LiteralPath 'path-to-downloaded-file'.
No profit outcome is promised. The experimental learner records PAPER shadow decisions; its predictions do not control order execution. Review evidence, charges and risk before drawing conclusions.
This website introduces TradeDrishti and provides an interactive demonstration. It is not connected to a market data feed or broker and cannot execute orders.
The desktop app offers optional local evidence review and experimental PAPER learning. This website does not run those models; its displayed checklist and market scenarios are illustrative.
No. It is designed around analysis and discipline. Trading can result in losses, and the demo scenarios are educational examples rather than recommendations.