{"slug": "claude-sonnet-5-5-an-80-cache-hit-rate-is-worth-about-as-much-as-the-batch-api", "title": "Claude Sonnet 5.5: an 80% cache-hit rate is worth about as much as the Batch API", "summary": "IteraCompute published a pricing analysis showing that an 80% prompt-cache hit rate on Claude Sonnet 5.5 costs about the same as Anthropic's Batch API discount, with a 100M-input/10M-output month totaling $148 at 80% cache hits in real time versus $150 on Batch with no caching. The analysis notes Sonnet 5.5 cache reads are priced at 0.05x base input rather than the 0.1x used by older Claude models, and that batch discounts stack on top of caching. IteraCompute is serving the model through an OpenAI-compatible API at Anthropic's Batch per-token prices without the queue.", "body_md": "Anthropic's Batch API halves the price of Claude Sonnet 5.5 if you can wait up to 24 hours. Prompt caching cuts the input side by 95% on every hit, in real time. On a typical month, the two end up close to the same bill, so it's worth knowing which lever you're actually pulling before you move traffic into a queue.\n\n**Anthropic list prices for Claude Sonnet 5.5 (USD per 1M tokens, checked 2026-11-05)**\n\n| Tier | Input | Cache read | Cache write (5m / 1h) | Output | \n|---|---|---|---|---|\n| Standard (real time) | $2.00 | $0.10 | $2.50 / $4.00 | $10.00 | \n| Batch API (async, up to 24h) | $1.00 | $0.05 | $1.25 / $2.00 | $5.00 | \n\nCache reads on Sonnet 5.5 are 0.05x of base input, not the 0.1x most older Claude models use, and the batch discount stacks on top. US-only inference (`inference_geo`) adds 1.1x to everything.\n\n**One month: 100M input tokens, 10M output tokens** (cache writes left out)\n\n| Cache-hit rate | Standard | Batch | \n|---|---|---|\n| 0% | $300.00 | $150.00 | \n| 50% | $205.00 | $102.50 | \n| 80% | $148.00 | $74.00 | \n| 90% | $129.00 | $64.50 | \n\nThe number to notice: **Batch at 0% cache hits ($150) costs about the same as real time at 80% hits ($148).** If your prompts share a long, stable prefix, fixing caching gets you the batch discount without the 24-hour wait. Batch only wins clearly when the prefix varies per request, or when you can get both.\n\n**Where the money goes once caching works**\n\nAt 90% hits on the standard tier, input is $29 of the $129 and output is $100. From that point, input price barely matters. What moves the bill:\n\n**A quick way to decide**\n\nThis is IteraCompute. We serve `anthropic/claude-sonnet-5-5` through an OpenAI-compatible API, real time, at $1.00 input / $0.05 cached input / $1.25 cache write (5m) / $5.00 output per 1M tokens (checked 2026-11-05). That matches Anthropic's Batch per-token prices without the queue: $74 for the 80%-hit month above, against $148 on Anthropic's standard tier. If your workload can wait for Anthropic's Batch API, that's the same price direct. The same key also serves Claude Haiku 5.5 and low-cost open models like DeepSeek V4.1 Flash, so a mixed route doesn't need a second client: [Iteracompute Claude Sonnet 5.5](https://iteracompute.com/models/claude-sonnet-5-5/?utm_source=dev&utm_medium=social&utm_campaign=one_api_mix_202611&utm_content=20261105_dev_sonnet_01)\n\nSources: Anthropic pricing docs (model pricing, batch processing, prompt caching, data residency sections), checked 2026-11-05.", "url": "https://wpnews.pro/news/claude-sonnet-5-5-an-80-cache-hit-rate-is-worth-about-as-much-as-the-batch-api", "canonical_source": "https://dev.to/iteracomputeai/claude-sonnet-55-an-80-cache-hit-rate-is-worth-about-as-much-as-the-batch-api-5bgi", "published_at": "2026-11-05 02:41:00+00:00", "updated_at": "2026-11-05 02:52:05.258075+00:00", "lang": "en", "topics": ["large-language-models", "ai-infrastructure", "ai-tools", "mlops"], "entities": ["Anthropic", "Claude Sonnet 5.5", "IteraCompute", "Claude Haiku 5.5", "DeepSeek V4.1 Flash"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/claude-sonnet-5-5-an-80-cache-hit-rate-is-worth-about-as-much-as-the-batch-api", "markdown": "https://wpnews.pro/news/claude-sonnet-5-5-an-80-cache-hit-rate-is-worth-about-as-much-as-the-batch-api.md", "text": "https://wpnews.pro/news/claude-sonnet-5-5-an-80-cache-hit-rate-is-worth-about-as-much-as-the-batch-api.txt", "jsonld": "https://wpnews.pro/news/claude-sonnet-5-5-an-80-cache-hit-rate-is-worth-about-as-much-as-the-batch-api.jsonld"}}