Cost mode:

Category: Structured Data & Fact Extraction · Rail: absolute · Typical I/O: 4084→5339 tokens

Models

Frontier on this task: Gemini 3.1 Pro Preview at 9.84 / 10. Quality bar at 90%: 8.86.

point-estimate floor (CI low) · upper CI (less certain) · Bars sorted by blended cost; best-value model first. Greyed rows are MEDIUM+ models whose point estimate clears the bar but whose CI low does not.

ModelQuality scoreCI lowCost / 1k runsvs best value
DeepSeek V4 Flash9.54 / 109.40$0.40best value
MiniMax M39.46 / 109.30$5.2913x more expensive
GPT-5.4 Nano9.46 / 109.22$7.6119x more expensive
DeepSeek V4 Pro9.54 / 109.40$11.5229x more expensive
GPT-5.6 Luna9.51 / 109.27$12.1530x more expensive
Qwen 3.7 Plus9.31 / 109.17$22.4756x more expensive
Qwen 3.6 Flash8.99 / 108.49$23.1358x more expensive
Grok 4.59.23 / 109.04$29.3073x more expensive
Qwen 3.6 Plus9.65 / 109.47$37.2593x more expensive
Gemini 3.5 Flash9.70 / 109.64$42.96107x more expensive
Kimi K2.69.71 / 109.61$55.73139x more expensive
Gemini 3.1 Pro Preview9.84 / 109.74$62.16155x more expensive
GPT-5.6 Sol9.56 / 109.32$62.47156x more expensive
GPT-5.58.96 / 108.55$205.70513x more expensive
Claude Haiku 4.58.28 / 107.90$30.6777x more expensive
Claude Sonnet 4.68.66 / 108.26$105.33263x more expensive
GPT-5.4 Mini7.94 / 107.52$18.4846x more expensive
Meta Muse Spark 1.18.80 / 108.33$38.9897x more expensive
GPT-5.6 Terra8.60 / 108.11$33.5884x more expensive

Cost breakdown

ModelQualityConfidenceCost / 1k runsOverpayMode
DeepSeek V4 Flash DeepSeek9.54 / 10 CI [9.40, 9.68]RANKED$0.40best valuebatch
MiniMax M3 MiniMax9.46 / 10 CI [9.30, 9.61]RANKED$5.2913xbatch
GPT-5.4 Nano OpenAI9.46 / 10 CI [9.22, 9.71]HIGH$7.6119xbatch
DeepSeek V4 Pro DeepSeek9.54 / 10 CI [9.40, 9.69]RANKED$11.5229xbatch
GPT-5.6 Luna OpenAI9.51 / 10 CI [9.27, 9.75]HIGH$12.1530xbatch
Qwen 3.7 Plus Alibaba Cloud (DashScope)9.31 / 10 CI [9.17, 9.44]RANKED$22.4756xbatch
Qwen 3.6 Flash Alibaba Cloud (DashScope)8.99 / 10 CI [8.49, 9.48]MEDIUM$23.1358xbatch
Grok 4.5 xAI9.23 / 10 CI [9.04, 9.42]RANKED$29.3073xbatch
Qwen 3.6 Plus Alibaba Cloud (DashScope)9.65 / 10 CI [9.47, 9.84]RANKED$37.2593xbatch
Gemini 3.5 Flash Gemini9.70 / 10 CI [9.64, 9.77]RANKED$42.96107xbatch
Kimi K2.6 Moonshot AI9.71 / 10 CI [9.61, 9.81]RANKED$55.73139xbatch
Gemini 3.1 Pro Preview best Gemini9.84 / 10 CI [9.74, 9.94]RANKED$62.16155xbatch
GPT-5.6 Sol OpenAI9.56 / 10 CI [9.32, 9.79]HIGH$62.47156xbatch
GPT-5.5 OpenAI8.96 / 10 CI [8.55, 9.36]MEDIUM$205.70513xbatch

Overpay shows how much more you pay than the best-value model that clears the quality bar (marked ★) — the best-value good-enough option. "16x" means you overpay 16× — 16× that reference for no quality benefit above the bar. Typical call shape for this task: 4084 input tokens → 5339 output tokens, EMA-tracked from production traffic. Cost is the observed, all-in $ per 1,000 task runs: each model's own measured usage on this task — output verbosity, thinking/reasoning tokens, cache reads and writes, and the spend on its billed failures — priced at current list rates and adjusted by the billing overhead we actually reconcile against provider invoices. Models that answer tersely cost what they actually cost; models that think at length pay for it. Not comparable to providers' advertised $/1M list rates — this is what running the task costs, not a per-token price.

Prompt templates

The system + user template pair used for this task.

STRUCTURED_OUTPUT_EXTRACTION_SYSTEM_PROMPT + STRUCTURED_OUTPUT_EXTRACTION_USER_PROMPT (1565 calls in window)

System prompt

You are a JSON extraction assistant. Extract structured data from the provided text that matches the given JSON schema. Output ONLY the structured data matching the schema.

User prompt

Extract the structured data from the following text that matches this JSON schema:

## JSON Schema:
```json
{schema_json}
```

## Text to extract from:
{raw_text}