{
  "@context": "https://schema.org",
  "@type": "Dataset",
  "@id": "https://requesty.ai/data/coding-agent-cache-hit-rate-apr-2026",
  "id": "coding-agent-cache-apr26",
  "slug": "coding-agent-cache-hit-rate-apr-2026",
  "title": "Prompt-cache hit rate by coding agent, April 2026",
  "shortTitle": "Agent cache hit rate",
  "topic": "agentic",
  "abstract": "Which coding agents use prompt caching most effectively? In April 2026, Claude Code led at 92% cache hit rate (cached_tokens / input_tokens), followed by OpenCode at 89%. Kilo Code sits at 46% with 62K avg input tokens. The gap is architectural: agents that maintain consistent context prefixes across sequential calls achieve dramatically higher cache reuse.",
  "whyItMatters": "Cache efficiency is the single biggest lever on coding agent economics. At 92% cache hit, Claude Code pays roughly $0.30 per million effective input tokens versus $3.00 list price. At 46%, Kilo Code pays $1.62 per million. That 5.4x cost difference compounds across every call in every session, enabling high-cache agents to sustain intensive workflows at fraction of the cost.",
  "questions": [
    "Which coding agent has the best prompt caching efficiency?",
    "How much does prompt caching reduce coding agent costs?",
    "How does Claude Code achieve 92% cache hit rate?"
  ],
  "period": "Apr 2026",
  "updated": "2026-05-16",
  "license": "CC BY 4.0",
  "licenseUrl": "https://creativecommons.org/licenses/by/4.0/",
  "caveats": [
    "Cache hit rate depends on both agent architecture and model provider. Anthropic, Bedrock, and Vertex have different caching implementations.",
    "Agents with very low traffic (Cursor, GitHub Copilot, Codex CLI) are excluded due to insufficient sample size."
  ],
  "keyFindings": [
    "Claude Code: 92% cache hit rate, the leader by a wide margin.",
    "OpenCode: 89%. Second only to Claude Code despite different architecture.",
    "Roo Code: 74%. Solid but significantly behind Claude Code.",
    "Kilo Code: 46%. Smaller context windows (62K vs 84K) reduce prefix reuse opportunity.",
    "Higher cache rates correlate strongly with lower per-call costs across all agents."
  ],
  "columns": [
    {
      "key": "agent",
      "label": "Agent",
      "unit": "count"
    },
    {
      "key": "cache_hit_rate",
      "label": "Cache hit rate",
      "unit": "percent"
    }
  ],
  "rows": [
    {
      "agent": "Claude Code",
      "cache_hit_rate": 0.919
    },
    {
      "agent": "OpenCode",
      "cache_hit_rate": 0.89
    },
    {
      "agent": "Aider",
      "cache_hit_rate": 0.84
    },
    {
      "agent": "Zed",
      "cache_hit_rate": 0.801
    },
    {
      "agent": "Roo Code",
      "cache_hit_rate": 0.736
    },
    {
      "agent": "Forge",
      "cache_hit_rate": 0.639
    },
    {
      "agent": "Cline",
      "cache_hit_rate": 0.614
    },
    {
      "agent": "Kilo Code",
      "cache_hit_rate": 0.455
    }
  ],
  "rowKey": "agent",
  "citation": {
    "apa": "Requesty (2026). Prompt-cache hit rate by coding agent, April 2026. Requesty Data. https://requesty.ai/data/coding-agent-cache-hit-rate-apr-2026",
    "bibtex": "@misc{requesty_coding_agent_cache_hit_rate_apr_2026,\n  author       = {{Requesty}},\n  title        = {Prompt-cache hit rate by coding agent, April 2026},\n  year         = {2026},\n  howpublished = {\\url{https://requesty.ai/data/coding-agent-cache-hit-rate-apr-2026}},\n  note         = {Requesty Data}\n}"
  },
  "permalink": "https://requesty.ai/data/coding-agent-cache-hit-rate-apr-2026",
  "downloads": {
    "json": "https://requesty.ai/data/coding-agent-cache-hit-rate-apr-2026/data.json",
    "csv": "https://requesty.ai/data/coding-agent-cache-hit-rate-apr-2026/data.csv",
    "markdown": "https://requesty.ai/data/coding-agent-cache-hit-rate-apr-2026.md"
  },
  "citedIn": [],
  "image": "https://requesty.ai/data/coding-agent-cache-hit-rate-apr-2026/opengraph-image",
  "source": {
    "organization": "Requesty",
    "url": "https://requesty.ai"
  }
}