@intuned/browser-dev 0.1.16-dev.0 → 0.1.17-dev.0

package/accumulate_llm_cost.py

@@ -0,0 +1,120 @@
+#!/usr/bin/env python3
+"""
+Accumulate total LLM cost (in cents) across all runs in a results JSONL file.
+
+For each line in the results JSONL:
+  - read the `log_url` field
+  - download that logs JSONL
+  - find every log line whose message contains "Total LLM Cost In Cents: <number>"
+  - sum all those numbers
+
+Usage:
+    python accumulate_llm_cost.py /path/to/results.jsonl
+    python accumulate_llm_cost.py /path/to/results.jsonl --workers 16
+"""
+
+import argparse
+import json
+import re
+import sys
+import urllib.request
+from concurrent.futures import ThreadPoolExecutor, as_completed
+
+# Matches: "Total LLM Cost In Cents: 01.1"  -> captures "01.1"
+COST_RE = re.compile(r"Total LLM Cost In Cents:\s*([0-9]*\.?[0-9]+)", re.IGNORECASE)
+
+
+def download_text(url: str, timeout: int = 120) -> str:
+    req = urllib.request.Request(url, headers={"User-Agent": "llm-cost-accumulator"})
+    with urllib.request.urlopen(req, timeout=timeout) as resp:
+        return resp.read().decode("utf-8", errors="replace")
+
+
+def costs_from_log_text(text: str):
+    """Yield every cost value found in a logs JSONL blob."""
+    for line in text.splitlines():
+        line = line.strip()
+        if not line:
+            continue
+        message = None
+        try:
+            obj = json.loads(line)
+            message = obj.get("message")
+        except json.JSONDecodeError:
+            # fall back to scanning the raw line
+            message = line
+        if not message:
+            continue
+        for m in COST_RE.finditer(str(message)):
+            yield float(m.group(1))
+
+
+def process_run(record: dict):
+    """Return (label, total_cents, hit_count, error) for one results record."""
+    api = record.get("apiInfo", {})
+    name = api.get("name", "?")
+    run_id = api.get("runId") or record.get("projectJobRun", {}).get("id", "?")
+    label = f"{name} / {run_id}"
+
+    log_url = api.get("log_url")
+    if not log_url:
+        return label, 0.0, 0, "no log_url"
+
+    try:
+        text = download_text(log_url)
+    except Exception as e:  # noqa: BLE001
+        return label, 0.0, 0, f"download failed: {e}"
+
+    costs = list(costs_from_log_text(text))
+    return label, sum(costs), len(costs), None
+
+
+def main():
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("results", help="Path to the results JSONL file")
+    parser.add_argument("--workers", type=int, default=8, help="Parallel downloads")
+    parser.add_argument("--quiet", action="store_true", help="Only print the grand total")
+    args = parser.parse_args()
+
+    with open(args.results, "r", encoding="utf-8") as f:
+        records = []
+        for ln, line in enumerate(f, 1):
+            line = line.strip()
+            if not line:
+                continue
+            try:
+                records.append(json.loads(line))
+            except json.JSONDecodeError as e:
+                print(f"  [warn] skipping malformed line {ln}: {e}", file=sys.stderr)
+
+    if not args.quiet:
+        print(f"Loaded {len(records)} run(s) from {args.results}\n")
+
+    grand_total = 0.0
+    total_hits = 0
+    errors = 0
+
+    with ThreadPoolExecutor(max_workers=args.workers) as pool:
+        futures = {pool.submit(process_run, rec): rec for rec in records}
+        for fut in as_completed(futures):
+            label, total, hits, err = fut.result()
+            if err:
+                errors += 1
+                if not args.quiet:
+                    print(f"  [error] {label}: {err}")
+                continue
+            grand_total += total
+            total_hits += hits
+            if not args.quiet:
+                print(f"  {label}: {total:.4f} cents ({hits} entr{'y' if hits == 1 else 'ies'})")
+
+    print("\n" + "=" * 60)
+    print(f"Runs processed : {len(records)}")
+    print(f"Cost entries   : {total_hits}")
+    print(f"Errors         : {errors}")
+    print(f"TOTAL LLM COST : {grand_total:.4f} cents  (${grand_total / 100:.4f})")
+    print("=" * 60)
+
+
+if __name__ == "__main__":
+    main()

package/dist/optimized-extractors/common/aiModelsValidations.js

@@ -3,30 +3,11 @@
 Object.defineProperty(exports, "__esModule", {
   value: true
 });
-exports.SUPPPORTED_GPT_MODELS = exports.SUPPPORTED_CLAUDE_MODELS = exports.SUPPORTED_MODELS = exports.MODELS_MAPPINGS = exports.MAX_TOKENS_OVERRIDES = exports.GPT_MODELS = exports.GOOGLE_MODELS = exports.CLAUDE_VISION_SUPPORTED_MODELS = exports.CLAUDE_MODELS_MAPPINGS = exports.CLAUDE_MODELS = void 0;
-const CLAUDE_MODELS = exports.CLAUDE_MODELS = ["claude-opus-4-20250514", "claude-sonnet-4-20250514", "claude-3-7-sonnet-20250219", "claude-3-5-sonnet-20241022", "claude-3-5-sonnet-20240620", "claude-3-5-haiku-20241022", "claude-3-opus-20240229", "claude-3-sonnet-20240229", "claude-3-haiku-20240307"];
+exports.MAX_TOKENS_OVERRIDES = void 0;
 const MAX_TOKENS_OVERRIDES = exports.MAX_TOKENS_OVERRIDES = {
   "claude-3-5-sonnet-20240620": 8192,
   "gemini-1.5-pro-002": 8192,
   "gemini-1.5-flash-8b-002": 8192,
   "gemini-1.5-flash-002": 8192,
   "gemini-2.0-flash-exp": 8192
-};
-const CLAUDE_VISION_SUPPORTED_MODELS = exports.CLAUDE_VISION_SUPPORTED_MODELS = [...CLAUDE_MODELS];
-const CLAUDE_MODELS_MAPPINGS = exports.CLAUDE_MODELS_MAPPINGS = {
-  "claude-3-haiku": "claude-3-haiku-20240307",
-  "claude-3-5-haiku": "claude-3-5-haiku-20241022",
-  "claude-3-opus": "claude-3-opus-20240229",
-  "claude-3-sonnet": "claude-3-sonnet-20240229",
-  "claude-3.5-sonnet": "claude-3-5-sonnet-20241022",
-  "claude-4-sonnet": "claude-sonnet-4-20250514",
-  "claude-4-opus": "claude-opus-4-20250514"
-};
-const GPT_MODELS = exports.GPT_MODELS = ["gpt-4.1", "gpt-4.1-mini", "gpt-4.1-nano", "gpt-4o", "gpt-4o-mini", "gpt-4o-audio-preview", "gpt-4-turbo", "gpt-4", "gpt-3.5-turbo", "o1", "o1-mini", "o1-preview", "o3-mini", "o3", "o4-mini", "chatgpt-4o-latest", "gpt4-turbo"];
-const GOOGLE_MODELS = exports.GOOGLE_MODELS = ["gemini-2.5-pro", "gemini-2.5-flash", "gemini-2.5-flash-lite", "gemini-2.5-flash-lite-preview-06-17", "gemini-2.0-flash", "gemini-1.5-pro", "gemini-1.5-pro-latest", "gemini-1.5-flash", "gemini-1.5-flash-latest", "gemini-1.5-flash-8b", "gemini-1.5-flash-8b-latest"];
-const MODELS_MAPPINGS = exports.MODELS_MAPPINGS = {
-  ...CLAUDE_MODELS_MAPPINGS
-};
-const SUPPPORTED_CLAUDE_MODELS = exports.SUPPPORTED_CLAUDE_MODELS = ["claude-3-5-haiku-20241022", "claude-3-5-haiku-latest", "claude-3-5-sonnet-20240620", "claude-3-5-sonnet-20241022", "claude-3-5-sonnet-latest", "claude-3-7-sonnet-20250219", "claude-3-7-sonnet-latest", "claude-3-haiku-20240307", "claude-4-opus-20250514", "claude-4-sonnet-20250514", "claude-opus-4-1", "claude-opus-4-1-20250805", "claude-opus-4-20250514", "claude-sonnet-4-20250514"];
-const SUPPPORTED_GPT_MODELS = exports.SUPPPORTED_GPT_MODELS = ["gpt-3.5-turbo", "gpt-3.5-turbo-0125", "gpt-3.5-turbo-0301", "gpt-3.5-turbo-0613", "gpt-3.5-turbo-1106", "gpt-3.5-turbo-16k", "gpt-3.5-turbo-16k-0613", "gpt-3.5-turbo-instruct", "gpt-3.5-turbo-instruct-0914", "gpt-4", "gpt-4-0314", "gpt-4-0613", "gpt-4-32k", "gpt-4-32k-0314", "gpt-4-32k-0613", "gpt-4-turbo", "gpt-4-turbo-2024-04-09", "gpt-4.1", "gpt-4.1-2025-04-14", "gpt-4.1-mini", "gpt-4.1-mini-2025-04-14", "gpt-4.1-nano", "gpt-4.1-nano-2025-04-14", "gpt-4o", "gpt-4o-2024-05-13", "gpt-4o-2024-08-06", "gpt-4o-2024-11-20", "gpt-4o-mini", "gpt-4o-mini-2024-07-18", "gpt-5", "gpt-5-2025-08-07", "gpt-5-chat", "gpt-5-chat-latest", "gpt-5-mini", "gpt-5-mini-2025-08-07", "gpt-5-nano", "gpt-5-nano-2025-08-07", "o1", "o1-2024-12-17", "o1-mini", "o1-mini-2024-09-12", "o1-pro", "o1-pro-2025-03-19", "o3", "o3-2025-04-16", "o3-deep-research", "o3-deep-research-2025-06-26", "o3-mini", "o3-mini-2025-01-31", "o3-pro", "o3-pro-2025-06-10", "o4-mini", "o4-mini-2025-04-16", "o4-mini-deep-research", "o4-mini-deep-research-2025-06-26"];
-const SUPPORTED_MODELS = exports.SUPPORTED_MODELS = [...SUPPPORTED_CLAUDE_MODELS, ...SUPPPORTED_GPT_MODELS];
+};

package/dist/optimized-extractors/common/extractStructuredDataUsingClaude.js

@@ -8,10 +8,10 @@ var _anthropicModel = require("../models/anthropicModel");
 var _neverthrow = require("neverthrow");
 var Errors = _interopRequireWildcard(require("../types/errors"));
 var _utils = require("./utils");
-var _Logger = require("../../common/Logger");
 var _aiModelsValidations = require("../common/aiModelsValidations");
 function _interopRequireWildcard(e, t) { if ("function" == typeof WeakMap) var r = new WeakMap(), n = new WeakMap(); return (_interopRequireWildcard = function (e, t) { if (!t && e && e.__esModule) return e; var o, i, f = { __proto__: null, default: e }; if (null === e || "object" != typeof e && "function" != typeof e) return f; if (o = t ? n : r) { if (o.has(e)) return o.get(e); o.set(e, f); } for (const t in e) "default" !== t && {}.hasOwnProperty.call(e, t) && ((i = (o = Object.defineProperty) && Object.getOwnPropertyDescriptor(e, t)) && (i.get || i.set) ? o(f, t, i) : f[t] = e[t]); return f; })(e, t); }
 async function extractStructuredDataUsingClaude(input) {
+  var _unwrappedResponse$us, _unwrappedResponse$us2;
   const {
     entityName,
     model,
@@ -56,8 +56,7 @@ async function extractStructuredDataUsingClaude(input) {
   const anthropic = (0, _anthropicModel.createAnthropicInstance)({
     apiKey
   });
-  const modelName = _aiModelsValidations.CLAUDE_MODELS_MAPPINGS[model] ?? model;
-  const maxTokens = _aiModelsValidations.MAX_TOKENS_OVERRIDES[modelName] ?? 4096;
+  const maxTokens = _aiModelsValidations.MAX_TOKENS_OVERRIDES[model] ?? 4096;
   const response = await (0, _neverthrow.fromPromise)(anthropic.messages.create({
     max_tokens: maxTokens,
     temperature: 0,
@@ -66,7 +65,7 @@ async function extractStructuredDataUsingClaude(input) {
       role: "user",
       content
     }],
-    model: modelName,
+    model,
     tools: [{
       input_schema: processedJsonSchema,
       name: toolName,
@@ -119,17 +118,11 @@ async function extractStructuredDataUsingClaude(input) {
   }
   const result = (0, _utils.getResultFromOutputSchema)(originalJsonSchema, entityName, tool.input);
   const callCost = response.value.response.headers.get("x-ai-cost-in-cents");
-  if (input.logAiCallCost) {
-    if (apiKey) {
-      _Logger.logger.info(`extractor ${input.identifier}: AI cost is not calculated (using custom API key)`);
-    } else if (callCost) {
-      const cost = parseFloat(callCost);
-      if (!isNaN(cost)) {
-        _Logger.logger.info(`extractor ${input.identifier}: AI cost is $${cost / 100}`);
-      }
-    }
-  }
+  const costInCents = (0, _utils.parseCostInCents)(callCost);
+  const totalTokens = (((_unwrappedResponse$us = unwrappedResponse.usage) === null || _unwrappedResponse$us === void 0 ? void 0 : _unwrappedResponse$us.input_tokens) ?? 0) + (((_unwrappedResponse$us2 = unwrappedResponse.usage) === null || _unwrappedResponse$us2 === void 0 ? void 0 : _unwrappedResponse$us2.output_tokens) ?? 0);
   return (0, _neverthrow.ok)({
-    result
+    result,
+    costInCents,
+    totalTokens
   });
 }

package/dist/optimized-extractors/common/extractStructuredDataUsingGoogle.js

@@ -15,10 +15,7 @@ async function extractStructuredDataUsingGoogle(input) {
   if (!input.apiKey) {
     return (0, _neverthrow.err)(Errors.invalidInput("Google AI is only supported with a custom API key. Please provide it or use a different AI provider."));
   }
-  let model = input.model;
-  if (input.model in _aiModelsValidations.MODELS_MAPPINGS) {
-    model = _aiModelsValidations.MODELS_MAPPINGS[input.model];
-  }
+  const model = input.model;
   const googleGenAi = (0, _google.createGoogleGenerativeAI)({
     apiKey: input.apiKey
   });

package/dist/optimized-extractors/common/extractStructuredDataUsingOpenAi.js

@@ -7,12 +7,10 @@ exports.extractStructuredDataUsingOpenAi = extractStructuredDataUsingOpenAi;
 var _neverthrow = require("neverthrow");
 var Errors = _interopRequireWildcard(require("../types/errors"));
 var _utils = require("./utils");
-var _Logger = require("../../common/Logger");
-var _aiModelsValidations = require("../common/aiModelsValidations");
 var _openaiModel = require("../models/openaiModel");
 function _interopRequireWildcard(e, t) { if ("function" == typeof WeakMap) var r = new WeakMap(), n = new WeakMap(); return (_interopRequireWildcard = function (e, t) { if (!t && e && e.__esModule) return e; var o, i, f = { __proto__: null, default: e }; if (null === e || "object" != typeof e && "function" != typeof e) return f; if (o = t ? n : r) { if (o.has(e)) return o.get(e); o.set(e, f); } for (const t in e) "default" !== t && {}.hasOwnProperty.call(e, t) && ((i = (o = Object.defineProperty) && Object.getOwnPropertyDescriptor(e, t)) && (i.get || i.set) ? o(f, t, i) : f[t] = e[t]); return f; })(e, t); }
 async function extractStructuredDataUsingOpenAi(input) {
-  var _completion$value$dat, _completion$value$dat2;
+  var _completion$value$dat, _completion$value$dat2, _completion$value$dat3;
   const {
     entityName,
     model,
@@ -50,14 +48,17 @@ async function extractStructuredDataUsingOpenAi(input) {
     }));
     content.push(...imageContent);
   }
-  const modelName = _aiModelsValidations.MODELS_MAPPINGS[input.model] ?? input.model;
+  const modelName = input.model;
+  const supportsCustomTemperature = !/^(o\d|gpt-5)/i.test(modelName);
   const toolName = `extract_${entityName}`;
   const openAiInstance = (0, _openaiModel.createOpenAIInstance)({
     apiKey
   });
   const completion = await (0, _neverthrow.fromPromise)(openAiInstance.chat.completions.create({
-    max_tokens: 4000,
-    temperature: 0,
+    max_completion_tokens: 4000,
+    ...(supportsCustomTemperature ? {
+      temperature: 0
+    } : {}),
     model: modelName,
     messages: [{
       role: "system",
@@ -116,17 +117,11 @@ async function extractStructuredDataUsingOpenAi(input) {
   const result = (0, _utils.getResultFromOutputSchema)(originalJsonSchema, entityName, parsedData.value);
   const formatted = (0, _utils.cleanupAiResult)(result);
   const callCost = completion.value.response.headers.get("x-ai-cost-in-cents");
-  if (input.logAiCallCost) {
-    if (apiKey) {
-      _Logger.logger.info(`extractor ${input.identifier}: AI cost is not calculated (using custom API key)`);
-    } else if (callCost) {
-      const cost = parseFloat(callCost);
-      if (!isNaN(cost)) {
-        _Logger.logger.info(`extractor ${input.identifier}: AI cost is $${cost / 100}`);
-      }
-    }
-  }
+  const costInCents = (0, _utils.parseCostInCents)(callCost);
+  const totalTokens = (_completion$value$dat3 = completion.value.data.usage) === null || _completion$value$dat3 === void 0 ? void 0 : _completion$value$dat3.total_tokens;
   return (0, _neverthrow.ok)({
-    result: formatted
+    result: formatted,
+    costInCents,
+    totalTokens
   });
 }

package/dist/optimized-extractors/common/extractStrucutredDataUsingAiInstance.js

@@ -7,11 +7,10 @@ exports.extractStructuredDataUsingAiInstance = extractStructuredDataUsingAiInsta
 var _neverthrow = require("neverthrow");
 var Errors = _interopRequireWildcard(require("../types/errors"));
 var _utils = require("./utils");
-var _Logger = require("../../common/Logger");
 var _ai = require("ai");
 function _interopRequireWildcard(e, t) { if ("function" == typeof WeakMap) var r = new WeakMap(), n = new WeakMap(); return (_interopRequireWildcard = function (e, t) { if (!t && e && e.__esModule) return e; var o, i, f = { __proto__: null, default: e }; if (null === e || "object" != typeof e && "function" != typeof e) return f; if (o = t ? n : r) { if (o.has(e)) return o.get(e); o.set(e, f); } for (const t in e) "default" !== t && {}.hasOwnProperty.call(e, t) && ((i = (o = Object.defineProperty) && Object.getOwnPropertyDescriptor(e, t)) && (i.get || i.set) ? o(f, t, i) : f[t] = e[t]); return f; })(e, t); }
 async function extractStructuredDataUsingAiInstance(input) {
-  var _apiResult$value$tool, _apiResult$value$tool2, _apiResult$value$resp;
+  var _apiResult$value$tool, _apiResult$value$tool2, _apiResult$value$resp, _apiResult$value$usag;
   const {
     entityName,
     model,
@@ -20,9 +19,7 @@ async function extractStructuredDataUsingAiInstance(input) {
     text,
     extraUserMessages,
     images,
-    apiKey,
-    apiName,
-    maxTokens
+    apiName
   } = input;
   const processedJsonSchema = (0, _utils.processInputSchema)(originalJsonSchema, entityName);
   const content = [];
@@ -106,17 +103,11 @@ async function extractStructuredDataUsingAiInstance(input) {
   const result = (0, _utils.getResultFromOutputSchema)(originalJsonSchema, entityName, extractedData);
   const formatted = (0, _utils.cleanupAiResult)(result);
   const callCost = (_apiResult$value$resp = apiResult.value.response.headers) === null || _apiResult$value$resp === void 0 ? void 0 : _apiResult$value$resp["x-ai-cost-in-cents"];
-  if (input.logAiCallCost) {
-    if (apiKey) {
-      _Logger.logger.info(`extractor ${input.identifier}: AI cost is not calculated (using custom API key)`);
-    } else if (callCost) {
-      const cost = parseFloat(callCost);
-      if (!isNaN(cost)) {
-        _Logger.logger.info(`extractor ${input.identifier}: AI cost is $${cost / 100}`);
-      }
-    }
-  }
+  const costInCents = (0, _utils.parseCostInCents)(callCost);
+  const totalTokens = (_apiResult$value$usag = apiResult.value.usage) === null || _apiResult$value$usag === void 0 ? void 0 : _apiResult$value$usag.totalTokens;
   return (0, _neverthrow.ok)({
-    result: formatted
+    result: formatted,
+    costInCents,
+    totalTokens
   });
 }

package/dist/optimized-extractors/common/findTableHeaders.js

@@ -10,7 +10,7 @@ var _imageSize = require("image-size");
 var _neverthrow = require("neverthrow");
 var Errors = _interopRequireWildcard(require("../types/errors"));
 function _interopRequireWildcard(e, t) { if ("function" == typeof WeakMap) var r = new WeakMap(), n = new WeakMap(); return (_interopRequireWildcard = function (e, t) { if (!t && e && e.__esModule) return e; var o, i, f = { __proto__: null, default: e }; if (null === e || "object" != typeof e && "function" != typeof e) return f; if (o = t ? n : r) { if (o.has(e)) return o.get(e); o.set(e, f); } for (const t in e) "default" !== t && {}.hasOwnProperty.call(e, t) && ((i = (o = Object.defineProperty) && Object.getOwnPropertyDescriptor(e, t)) && (i.get || i.set) ? o(f, t, i) : f[t] = e[t]); return f; })(e, t); }
-async function getTableHeadersUsingAi(handle, identifier) {
+async function getTableHeadersUsingAi(handle, model) {
   var _response$error;
   let image;
   try {
@@ -57,7 +57,7 @@ async function getTableHeadersUsingAi(handle, identifier) {
         }
       }]
     }],
-    model: "claude-3-haiku-20240307",
+    model,
     tools: [{
       input_schema: {
         type: "object",

package/dist/optimized-extractors/common/index.js

@@ -13,6 +13,7 @@ var _extractStructuredDataUsingOpenAi = require("./extractStructuredDataUsingOpe
 var _utils = require("./utils");
 var _extractStructuredDataUsingGoogle = require("./extractStructuredDataUsingGoogle");
 var _getModelProvider = require("../../common/getModelProvider");
+var _Logger = require("../../common/Logger");
 function isClaudeModel(model) {
   return (0, _getModelProvider.getModelProvider)(model) === "anthropic";
 }
@@ -47,9 +48,18 @@ async function extractStructuredDataUsingAi(input) {
     return (0, _neverthrow.err)(extractionResult.error);
   }
   const {
-    result
+    result,
+    costInCents,
+    totalTokens
   } = extractionResult.value;
+  if (costInCents !== undefined) {
+    _Logger.logger.info(`Total LLM Cost In Cents: ${costInCents}`);
+  } else if (totalTokens !== undefined) {
+    _Logger.logger.info(`Total LLM Tokens: ${totalTokens}`);
+  }
   return (0, _neverthrow.ok)({
-    result: (0, _utils.cleanupAiResult)(result)
+    result: (0, _utils.cleanupAiResult)(result),
+    costInCents,
+    totalTokens
   });
 }

package/dist/optimized-extractors/common/isTableHeaderOrFooter.js

@@ -9,7 +9,7 @@ var _zod = require("zod");
 var _neverthrow = require("neverthrow");
 var Errors = _interopRequireWildcard(require("../types/errors"));
 function _interopRequireWildcard(e, t) { if ("function" == typeof WeakMap) var r = new WeakMap(), n = new WeakMap(); return (_interopRequireWildcard = function (e, t) { if (!t && e && e.__esModule) return e; var o, i, f = { __proto__: null, default: e }; if (null === e || "object" != typeof e && "function" != typeof e) return f; if (o = t ? n : r) { if (o.has(e)) return o.get(e); o.set(e, f); } for (const t in e) "default" !== t && {}.hasOwnProperty.call(e, t) && ((i = (o = Object.defineProperty) && Object.getOwnPropertyDescriptor(e, t)) && (i.get || i.set) ? o(f, t, i) : f[t] = e[t]); return f; })(e, t); }
-async function isTableHeaderOrFooter(content) {
+async function isTableHeaderOrFooter(content, model) {
   var _response$error;
   if (!content) {
     return (0, _neverthrow.ok)({
@@ -39,7 +39,7 @@ async function isTableHeaderOrFooter(content) {
             a table header is a row that contains labels for table columns, and footer usually has pagination information or summary of the table`
       }, itemContent]
     }],
-    model: "claude-3-haiku-20240307",
+    model,
     tools: [{
       input_schema: {
         type: "object",

package/dist/optimized-extractors/common/modelStringSupport.test.js

@@ -47,7 +47,7 @@ var _index = require("./index");
     (0, _vitest.expect)(result.success).toBe(true);
     if (result.success) {
       (0, _vitest.expect)(result.data).toEqual({
-        model: "claude-3-haiku",
+        model: "claude-haiku-4-5-20251001",
         type: "HTML"
       });
     }

package/dist/optimized-extractors/common/utils.js

@@ -6,11 +6,17 @@ Object.defineProperty(exports, "__esModule", {
 exports.cleanupAiResult = cleanupAiResult;
 exports.getRandomItems = getRandomItems;
 exports.getResultFromOutputSchema = getResultFromOutputSchema;
+exports.parseCostInCents = parseCostInCents;
 exports.processInputSchema = processInputSchema;
 function getRandomItems(arr, numItems) {
   const shuffled = arr.sort(() => 0.5 - Math.random());
   return shuffled.slice(0, numItems);
 }
+function parseCostInCents(headerValue) {
+  if (!headerValue) return undefined;
+  const cost = parseFloat(headerValue);
+  return isNaN(cost) ? undefined : cost;
+}
 function processInputSchema(originalJsonSchema, entityName) {
   const internalSchema = structuredClone(originalJsonSchema);
   delete internalSchema.description;

package/dist/optimized-extractors/listExtractionHelpers/runAiExtraction.js

@@ -209,7 +209,7 @@ async function splitDomAndExtractData({
     tableLocater
   } = await (0, _tablesUtils.isListTable)(listItemsContainerLocator, itemsSimplifiedHtml);
   const tableAsJsonArray = isTable ? await (0, _tablesUtils.createJsonFromTable)(pageAndSearchRegion.page) : [];
-  const tableHeaders = tableLocater ? await (0, _findTableHeaders.getTableHeadersUsingAi)(tableLocater, identifier) : undefined;
+  const tableHeaders = tableLocater ? await (0, _findTableHeaders.getTableHeadersUsingAi)(tableLocater, strategy.model) : undefined;
   if (tableHeaders && tableHeaders.isErr()) {
     return (0, _neverthrow.err)(tableHeaders.error);
   }

package/dist/optimized-extractors/listExtractionHelpers/utils/extractPropertiesUsingGPTFromArray.js

@@ -44,7 +44,7 @@ async function extractPropertiesUsingGPT({
   }
   if (possibleTableHeaderOrFooter) {
     const content = text ?? image;
-    const isHeader = await (0, _isTableHeaderOrFooter.isTableHeaderOrFooter)(content);
+    const isHeader = await (0, _isTableHeaderOrFooter.isTableHeaderOrFooter)(content, strategy.model);
     if (isHeader.isErr()) {
       return (0, _neverthrow.err)(isHeader.error);
     }
@@ -146,7 +146,7 @@ async function extractPropertiesWithHTMLStrategy({
   apiKey
 }) {
   const shouldUseTableData = !!tableAsJsonArray && tableAsJsonArray.length === items.length;
-  const isWeakModel = strategy.model === "claude-3-haiku" || strategy.model == "gpt3.5-turbo";
+  const isWeakModel = strategy.model.includes("haiku") || strategy.model.includes("turbo");
   const averageItemLength = items.reduce((sum, item) => {
     if (item.type !== "text") return sum;
     return sum + (0, _extractionHelpers.compressStringSpaces)(item.text).length;

package/dist/optimized-extractors/validators.js

@@ -113,7 +113,7 @@ const strategySchema = exports.strategySchema = _zod.z.union([htmlStrategySchema
     };
   }
 }).optional().default({
-  model: "claude-3-haiku",
+  model: "claude-haiku-4-5-20251001",
   type: "HTML"
 });
 const labelSchema = _zod.z.string({

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@intuned/browser-dev",
-  "version": "0.1.16-dev.0",
+  "version": "0.1.17-dev.0",
   "description": "runner package for intuned functions",
   "types": "./dist/index.d.ts",
   "typesVersions": {

package/dist/optimized-extractors/types/aiModelsValidation.js

@@ -1,45 +0,0 @@
-"use strict";
-
-Object.defineProperty(exports, "__esModule", {
-  value: true
-});
-exports.SUPPORTED_VISION_MODELS = exports.SUPPORTED_TEXT_MODELS = exports.SUPPORTED_GPT_MODELS = exports.SUPPORTED_GOOGLE_MODELS = exports.SUPPORTED_CLAUDE_MODELS = exports.MODELS_MAPPINGS = exports.MAX_TOKENS_OVERRIDES = exports.GPT_MODELS_MAPPINGS = exports.GOOGLE_MODELS_MAPPINGS = exports.CLAUDE_VISION_SUPPORTED_MODELS = exports.CLAUDE_ONLY_TEXT_MODELS = exports.CLAUDE_MODELS_MAPPINGS = void 0;
-const CLAUDE_ONLY_TEXT_MODELS = exports.CLAUDE_ONLY_TEXT_MODELS = ["claude-3-5-haiku", "claude-3-5-haiku-20241022"];
-const CLAUDE_VISION_SUPPORTED_MODELS = exports.CLAUDE_VISION_SUPPORTED_MODELS = ["claude-3-haiku", "claude-3-haiku-20240307", "claude-3.5-sonnet", "claude-3-5-sonnet-20240620", "claude-3-5-sonnet-20241022", "claude-opus-4", "claude-opus-4-20250514", "claude-sonnet-4", "claude-sonnet-4-20250514"];
-const SUPPORTED_CLAUDE_MODELS = exports.SUPPORTED_CLAUDE_MODELS = [...CLAUDE_ONLY_TEXT_MODELS, ...CLAUDE_VISION_SUPPORTED_MODELS];
-const CLAUDE_MODELS_MAPPINGS = exports.CLAUDE_MODELS_MAPPINGS = {
-  "claude-3-haiku": "claude-3-haiku-20240307",
-  "claude-3-5-haiku": "claude-3-5-haiku-20241022",
-  "claude-3.5-sonnet": "claude-3-5-sonnet-20241022",
-  "claude-opus-4": "claude-opus-4-20250514",
-  "claude-sonnet-4": "claude-sonnet-4-20250514"
-};
-const GPT_ONLY_TEXT_GPT_MODELS = ["gpt3.5-turbo", "gpt-3.5-turbo-0125"];
-const GPT_VISION_SUPPORTED_MODELS = ["gpt4-turbo", "gpt-4-turbo-2024-04-09", "gpt-4o", "gpt-4o-2024-05-13", "gpt-4o-mini", "gpt-4o-mini-2024-07-18"];
-const SUPPORTED_GPT_MODELS = exports.SUPPORTED_GPT_MODELS = [...GPT_ONLY_TEXT_GPT_MODELS, ...GPT_VISION_SUPPORTED_MODELS];
-const GPT_MODELS_MAPPINGS = exports.GPT_MODELS_MAPPINGS = {
-  "gpt4-turbo": "gpt-4-turbo-2024-04-09",
-  "gpt3.5-turbo": "gpt-3.5-turbo-0125",
-  "gpt-4o": "gpt-4o-2024-05-13",
-  "gpt-4o-mini": "gpt-4o-mini-2024-07-18"
-};
-const SUPPORTED_GOOGLE_MODELS = exports.SUPPORTED_GOOGLE_MODELS = ["gemini-1.5-pro", "gemini-1.5-pro-002", "gemini-1.5-flash-8b", "gemini-1.5-flash-8b-002", "gemini-1.5-flash", "gemini-1.5-flash-002", "gemini-2.0-flash-exp"];
-const GOOGLE_MODELS_MAPPINGS = exports.GOOGLE_MODELS_MAPPINGS = {
-  "gemini-1.5-pro": "gemini-1.5-pro-002",
-  "gemini-1.5-flash-8b": "gemini-1.5-flash-8b-002",
-  "gemini-1.5-flash": "gemini-1.5-flash-002"
-};
-const SUPPORTED_TEXT_MODELS = exports.SUPPORTED_TEXT_MODELS = [...SUPPORTED_CLAUDE_MODELS, ...SUPPORTED_GPT_MODELS, ...SUPPORTED_GOOGLE_MODELS];
-const SUPPORTED_VISION_MODELS = exports.SUPPORTED_VISION_MODELS = [...CLAUDE_VISION_SUPPORTED_MODELS, ...GPT_VISION_SUPPORTED_MODELS, ...SUPPORTED_GOOGLE_MODELS];
-const MODELS_MAPPINGS = exports.MODELS_MAPPINGS = {
-  ...GPT_MODELS_MAPPINGS,
-  ...CLAUDE_MODELS_MAPPINGS,
-  ...GOOGLE_MODELS_MAPPINGS
-};
-const MAX_TOKENS_OVERRIDES = exports.MAX_TOKENS_OVERRIDES = {
-  "claude-3-5-sonnet-20240620": 8192,
-  "gemini-1.5-pro-002": 8192,
-  "gemini-1.5-flash-8b-002": 8192,
-  "gemini-1.5-flash-002": 8192,
-  "gemini-2.0-flash-exp": 8192
-};