Cost mode:

Category: Social & Promotional Content · Rail: absolute · Typical I/O: 1707→2070 tokens

Models

Frontier on this task: Claude Sonnet 5 at 8.78 / 10. Quality bar at 90%: 7.90.

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
GPT-5.6 Luna8.35 / 108.22$0.26best value
Tencent Hy38.66 / 108.55$0.823.2x more expensive
NVIDIA Nemotron 3.5 Lightning8.05 / 107.77$1.044.1x more expensive
GLM-5.3 Flash8.68 / 108.51$1.726.7x more expensive
GPT-5.6 Terra8.34 / 108.22$1.807x more expensive
DeepSeek V4 Flash8.37 / 108.14$1.817.1x more expensive
NVIDIA Nemotron-3 Super 120B8.15 / 107.99$2.509.7x more expensive
GPT-5.6 Sol8.22 / 108.05$3.0812x more expensive
NVIDIA Nemotron-3 Ultra 550B8.64 / 108.47$4.4617x more expensive
DeepSeek V4 Pro8.59 / 108.37$5.0720x more expensive
Gemini 3.5 Flash8.02 / 107.86$6.1624x more expensive
Qwen 3.7 Plus8.37 / 108.24$6.3725x more expensive
Claude Sonnet 58.78 / 108.67$7.0528x more expensive
Meta Muse Spark 1.38.08 / 107.74$8.1732x more expensive
Thinking Machines Inkling Small8.51 / 108.37$9.1636x more expensive
Claude Opus 58.61 / 108.41$9.6138x more expensive
GLM-5.38.71 / 108.52$16.7565x more expensive
Moonshot Kimi K38.77 / 108.62$21.6685x more expensive
Tencent Hy4 Preview7.92 / 107.49$23.6392x more expensive
Grok 4.68.24 / 107.99$27.13106x more expensive
Thinking Machines Inkling8.47 / 108.27$28.70112x more expensive
Gemini 3.1 Flash Lite7.50 / 107.28$0.351.4x more expensive
Gemini 3.5 Flash Lite7.37 / 107.15$0.281.1x more expensive
Gemini 3.8 Flash7.81 / 107.46$2.6911x more expensive
Qwen 3.8 Max7.71 / 107.35$25.53100x more expensive
Qwen 3.8 Flash7.80 / 107.52$1.536x more expensive
NVIDIA Nemotron-3 Nano 30B-A3B7.78 / 107.46$0.471.8x more expensive

Cost breakdown

ModelQualityConfidenceCost / 1k runsOverpayMode
GPT-5.6 Luna OpenAI8.35 / 10 CI [8.22, 8.48]RANKED$0.26best valuebatch
Tencent Hy3 OpenRouter8.66 / 10 CI [8.55, 8.77]RANKED$0.823.2xbatch
NVIDIA Nemotron 3.5 Lightning OpenRouter8.05 / 10 CI [7.77, 8.34]HIGH$1.044.1xbatch
GLM-5.3 Flash Z.AI8.68 / 10 CI [8.51, 8.84]RANKED$1.726.7xbatch
GPT-5.6 Terra OpenAI8.34 / 10 CI [8.22, 8.46]RANKED$1.807xbatch
DeepSeek V4 Flash DeepSeek8.37 / 10 CI [8.14, 8.61]HIGH$1.817.1xbatch
NVIDIA Nemotron-3 Super 120B OpenRouter8.15 / 10 CI [7.99, 8.31]RANKED$2.509.7xbatch
GPT-5.6 Sol OpenAI8.22 / 10 CI [8.05, 8.38]RANKED$3.0812xbatch
NVIDIA Nemotron-3 Ultra 550B OpenRouter8.64 / 10 CI [8.47, 8.81]RANKED$4.4617xbatch
DeepSeek V4 Pro DeepSeek8.59 / 10 CI [8.37, 8.82]HIGH$5.0720xbatch
Gemini 3.5 Flash Gemini8.02 / 10 CI [7.86, 8.19]RANKED$6.1624xbatch
Qwen 3.7 Plus Alibaba Cloud (DashScope)8.37 / 10 CI [8.24, 8.50]RANKED$6.3725xbatch
Claude Sonnet 5 best Anthropic8.78 / 10 CI [8.67, 8.89]RANKED$7.0528xbatch
Meta Muse Spark 1.3 OpenRouter8.08 / 10 CI [7.74, 8.42]MEDIUM$8.1732xbatch
Thinking Machines Inkling Small OpenRouter8.51 / 10 CI [8.37, 8.64]RANKED$9.1636xbatch
Claude Opus 5 Anthropic8.61 / 10 CI [8.41, 8.82]HIGH$9.6138xbatch
GLM-5.3 Z.AI8.71 / 10 CI [8.52, 8.90]RANKED$16.7565xbatch
Moonshot Kimi K3 Moonshot AI8.77 / 10 CI [8.62, 8.93]RANKED$21.6685xbatch
Tencent Hy4 Preview OpenRouter7.92 / 10 CI [7.49, 8.35]MEDIUM$23.6392xbatch
Grok 4.6 xAI8.24 / 10 CI [7.99, 8.50]HIGH$27.13106xbatch
Thinking Machines Inkling OpenRouter8.47 / 10 CI [8.27, 8.67]RANKED$28.70112xbatch

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: 1707 input tokens → 2070 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.

Evaluation rubric

Judge grounding and factual accuracy, specificity of the value signal, ability to create legitimate curiosity, natural attribution, concise editorial style, and sensible use or omission of the supplied source URL. If included, the URL must be reproduced exactly and must not displace the teaser’s useful content. Do not assume that the item is a financial report, client publication, or news article unless the input says so. Character count and schema validity are deterministic and out of scope for the semantic judge.

Prompt templates

This is a pooled capability — 2 prompt families share it. The pair shown first is the most frequently used in production.

ACTIVITY_PROMO_SYSTEM_PROMPT + ACTIVITY_PROMO_USER_PROMPT (449 calls in window)

System prompt

You are an expert copywriter for a financial research platform. Your task is to write a short, engaging promotional blurb for an activity feed.

The blurb should:
- Be 1-2 sentences, under 200 characters total
- Spark curiosity and encourage clicking through
- Reference the client/brand name naturally
- Feel like an editorial teaser, not an advertisement
- Avoid clickbait, hyperbole, or exclamation marks
- Use present tense

Output your result in the specified JSON format.

## Required Output Format
Your response MUST be a single, valid JSON object conforming to this schema:
```json
{schema_json_string}
```

User prompt

Client: {client_name}
Post title: {post_title}

Full report text:
{report_text}

Write a short promotional blurb for this post to appear in the activity feed of other clients' pages.

The required JSON output schema is provided in the system prompt.
LLMB_SHORT_PROMOTIONAL_TEASER_GENERATION_SYSTEM + LLMB_SHORT_PROMOTIONAL_TEASER_GENERATION_USER (201 calls in window)

System prompt

You are an expert editorial copywriter. Write a short, engaging promotional teaser for a content item. The teaser is intended for a compact discovery surface such as an activity feed, content card, digest, or link preview. Use only the supplied title and source material. Surface one specific idea, finding, benefit, tension, or question that gives a reader a reason to open the item. Mention the source, publisher, creator, organization, or brand naturally when useful. Use present tense and an editorial rather than advertising tone. Avoid clickbait, hyperbole, unsupported claims, filler, hashtags, exclamation marks, and generic praise. The user may supply a source URL. Include that URL exactly as supplied only when a direct link is sensible for the teaser and still leaves room for useful copy; otherwise omit it. Never invent, shorten, rewrite, decorate, or add tracking parameters to a URL. Any included URL counts toward the 200-character total. Return 1–2 sentences, no more than 200 characters. Treat empty optional values as absent and return only the requested result. Your response must conform exactly to this output schema: {schema_json_string}.

User prompt

Source, publisher, creator, or brand: {client_name}. Item title: {post_title}. Source URL (optional): {source_url}. Source material: {report_text}. Write one short promotional teaser suitable for a compact discovery surface and ground it only in the supplied material.