Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
Python agentaianthropicclaude-codecompressioncontext-engineeringcontext-windowcursorfastapilangchainllmmcpopenaiprompt-engineeringproxypythonragtoken-optimizationtokenstypescript
Commits on main
fix(cache): stop prefix transforms from busting the provider cache (#3735)
a2bf5ed Tejas Chopra committed 1 month ago
fix(savings): report negative savings instead of flooring them at zero (#3744)
1455f00 Tejas Chopra committed 1 month ago
fix(install): apply managed env vars added after a deployment was installed (#3740)
a4cb2bc Tejas Chopra committed 1 month ago
fix(binaries): fail closed when a download carries no sha256 pin (#3724)
26a2c49 Tejas Chopra committed 1 month ago
fix(tokenizers): disable trust_remote_code and allowlist HF tokenizer repos (#3723)
34da21e Tejas Chopra committed 1 month ago
fix(tool-search): don't orphan deferred tools, and stop overstating the saving (#3739)
85f9e01 Tejas Chopra committed 1 month ago
feat(savings): carry exact cache-read cost in history rollups (#3734)
1cb779e gglucass committed 1 month ago
docs(pricing): correct MODEL_ALIASES comments that still name the pruned target (#3733)
d05bacb Tejas Chopra committed 1 month ago
fix(tests): ask litellm for a model it prices instead of naming a retired one (#3722)
0ad5e68 Tejas Chopra committed 1 month ago
fix(metrics): split rate-limited by source and label failures by provider (#3707)
5ff4ea1 Tejas Chopra committed 1 month ago
fix(pricing): price savings against the cache mix, not flat list price (#3699)
89a58fd Tejas Chopra committed 1 month ago
fix(stats): stop rejected upstream turns from feeding the savings funnel (#3615)
85fac8c gglucass committed 2 months ago
fix(proxy): keep OpenCode Zen /responses requests streaming (#3659)
6f404de Alvaro Cavalcanti committed 2 months ago
fix(sdk): emit the final SSE event when the stream has no trailing newline (#3581)
ebc0b36 Shubham Srivastava committed 2 months ago
fix(spreadsheet): render an .xls cell the way the .xlsx loader renders it (#3616)
261796f Lukas committed 2 months ago
fix: preserve JSON object fields in SmartCrusher (#3648)
545f544 Matt Van Horn committed 2 months ago
fix(tabular): pass a table through when one cell is past the csv field limit (#3618)
334317d Lukas committed 2 months ago
fix(proxy): enforce budget limits on OpenAI and Gemini routes (#3579)
f734c57 Johnny Wilson Dougherty committed 2 months ago
fix(learn): prevent Windows CRLF accumulation in context writers (#3594)
3c1a901 Trîggered committed 2 months ago
fix(learn): prevent Windows CRLF accumulation in context writers (#3594)
abde3e6 Trîggered committed 2 months ago
deps: bump the npm-minor-patch group across 3 directories with 6 updates (#3622)
2d5909c dependabot[bot] committed 2 months ago
deps: bump the cargo-minor-patch group across 1 directory with 5 updates (#3642)
e326dd7 dependabot[bot] committed 2 months ago
feat(router): elide dense machine-generated lines no compressor can shrink (#3685)
9263b42 gglucass committed 2 months ago
fix(kompress): bound download retry backoff before overflow (#3630)
a7858fd prakx committed 2 months ago
fix(proxy): stop get_recent deep-copying dropped message payloads (#3613)
44b8e7b gglucass committed 2 months ago
feat(beacon): schema v2 — routing/compression training signal, gzip transport (#3253)
67eb910 Tejas Chopra committed 2 months ago
fix(deps): bump anyio to 4.14.2 for CVE-2026-63374 and CVE-2026-64847 (#3662)
bc21c93 Tejas Chopra committed 2 months ago
feat(proxy): accept the OpenAI Responses shape on the gateway turn (#3661)
be00a79 Tejas Chopra committed 2 months ago