Guides / Cookbooks

Cookbooks · Techniques · Beginner · about 5 min

Ask many questions in one request

Put every question you have about a symbol into one request: forecasts and rules together, one round trip, and a fraction of the tokens of separate calls.

1. When to use it

   Whenever you ask more than one thing about the same state. Each request carries a fixed overhead (the state, the model's setup), so splitting questions across requests repeats it.

2. The request

   Three forecasts, two English "what is true now" questions and three rules, on SPY:

   {
     "model": "ora-fin1",
     "state": {
       "symbol": "SPY"
     },
     "questions": {
       "up_week": {
         "type": "yes_no",
         "instructions": "Will the stock be higher in a week?"
       },
       "drop5": {
         "type": "yes_no",
         "instructions": "Will it drop at least 5% at some point in the next month?"
       },
       "rally5": {
         "type": "yes_no",
         "instructions": "Will it rally at least 5% at some point in the next month?"
       },
       "above_200d": {
         "type": "yes_no",
         "instructions": "Is it above its 200-day moving average?"
       },
       "above_50d": {
         "type": "yes_no",
         "instructions": "Is it above its 50-day moving average?"
       },
       "overbought": {
         "type": "yes_no",
         "ask": "rsi(C,14) > 70"
       },
       "near_high": {
         "type": "yes_no",
         "ask": "C >= 0.98 * highest(H,250)"
       },
       "busy": {
         "type": "yes_no",
         "ask": "V > avg(V,20)"
       }
     }
   }

   Open this in the Playground

3. The real response

   {
     "model": "ora-fin1",
     "answers": {
       "up_week": {
         "type": "yes_no",
         "yes_no": 0.598
       },
       "drop5": {
         "type": "yes_no",
         "yes_no": 0.203
       },
       "rally5": {
         "type": "yes_no",
         "yes_no": 0.229
       },
       "above_200d": {
         "type": "yes_no",
         "yes_no": 0.9451,
         "rule": "C > sma(C, 200)",
         "basis": "comparison",
         "explain": "today's close > sma200"
       },
       "above_50d": {
         "type": "yes_no",
         "yes_no": 0.1128,
         "rule": "C > sma(C, 50)",
         "basis": "comparison",
         "explain": "today's close > sma50"
       },
       "overbought": {
         "type": "yes_no",
         "yes_no": 0.0537,
         "rule": "rsi(C,14) > 70",
         "basis": "rule"
       },
       "near_high": {
         "type": "yes_no",
         "yes_no": 0.0854,
         "rule": "C >= 0.98 * highest(H,250)",
         "basis": "rule"
       },
       "busy": {
         "type": "yes_no",
         "yes_no": 0.9463,
         "rule": "V > avg(V,20)",
         "basis": "rule"
       }
     },
     "usage": {
       "input_tokens": 516,
       "output_tokens": 32
     },
     "context": {
       "source": "server",
       "symbol": "SPY",
       "as_of": "2026-10-30"
     },
     "version": "ora-fin-0.1.1"
   }

   For comparison, up_week asked on its own:

   {
     "model": "ora-fin1",
     "answers": {
       "up_week": {
         "type": "yes_no",
         "yes_no": 0.598
       }
     },
     "usage": {
       "input_tokens": 290,
       "output_tokens": 4
     },
     "context": {
       "source": "server",
       "symbol": "SPY",
       "as_of": "2026-10-30"
     },
     "version": "ora-fin-0.1.1"
   }

4. The numbers

   Eight questions in one request used 516 input tokens. One question alone used 290, so eight separate requests would be about 8 × 290 = 2320, roughly 4 times as many, and eight round trips instead of one. The shared question came back the same both ways (up_week 0.598 and 0.598): answers don't depend on what else is asked.

5. The code

   cURLPythonJavaScriptPHPGoJavaC#RubyCopy

   Requires curl 7.76+. Save as run.sh, then: bash run.sh

   curl -sS https://oramodel.com/v1/systemone \
     -H "Authorization: Bearer $JEV_API_KEY" \
     -H "Content-Type: application/json" \
     -d '{
     "model": "ora-fin1",
     "state": {
       "symbol": "SPY"
     },
     "questions": {
       "up_week": {
         "type": "yes_no",
         "instructions": "Will the stock be higher in a week?"
       },
       "drop5": {
         "type": "yes_no",
         "instructions": "Will it drop at least 5% at some point in the next month?"
       },
       "rally5": {
         "type": "yes_no",
         "instructions": "Will it rally at least 5% at some point in the next month?"
       },
       "above_200d": {
         "type": "yes_no",
         "instructions": "Is it above its 200-day moving average?"
       },
       "above_50d": {
         "type": "yes_no",
         "instructions": "Is it above its 50-day moving average?"
       },
       "overbought": {
         "type": "yes_no",
         "ask": "rsi(C,14) > 70"
       },
       "near_high": {
         "type": "yes_no",
         "ask": "C >= 0.98 * highest(H,250)"
       },
       "busy": {
         "type": "yes_no",
         "ask": "V > avg(V,20)"
       }
     }
   }'

6. Variations

   • Name questions for what you will do with them (drop5, above_200d): the names are what action rules refer to.
   • One request per symbol, many questions each, many symbols in parallel: Rank a watchlist.
   • Rules and forecasts mix freely; rules need state.symbol.

Information, not investment advice. Every number on this page is a real response for the session shown; answers change with each session.

What's next

• Self-consistency: three wordings