🪢 Open source agent evals & observability: Trace, evaluate, and improve LLM applications with one open platform.
TypeScript analyticsautogenevaluationlangchainlarge-language-modelsllama-indexllmllm-evaluationllm-observabilityllmopsmonitoringobservabilityopen-sourceopenaiplaygroundprompt-engineeringprompt-managementself-hostedycombinator
Commits on main
fix(web): annotation-mode trace tags use the header's tag row spacing (#18520)
d4c85ba Trang Nguyen committed 1 month ago
feat(topics): add adaptive map exploration and compact workspace (#18492)
ab8afeb Nikita Kabardin committed 1 month ago
chore(clickhouse): emit unsampled per-query cost distributions (#18486)
3692810 Niklas Semmler committed 1 month ago
perf(scores): route score filter-option reads to the read replica (#18506)
42d3a5d David Traina committed 1 month ago
perf(sessions): route session score reads to the read replica (#18508)
ccaef3f David Traina committed 1 month ago
perf(datasets): route dataset run and experiment score reads to the read replica (#18499)
d994d17 David Traina committed 1 month ago
perf(scores): default trace score lookups to the read replica (#18509)
06e2fbd David Traina committed 1 month ago
perf(scores): route score metadata lookups to the read replica (#18505)
389f393 David Traina committed 1 month ago
perf(scores): route the scores table and count reads to the read replica (#18507)
2e1c4cf David Traina committed 1 month ago
perf(prompts): route prompt version metrics to the read replica (#18503)
82361f2 David Traina committed 1 month ago
perf(blob-export): route scores and traces blob storage exports to the read replica (#18497)
446ca9d David Traina committed 1 month ago
perf(scores): route the has-any-score UI check to the read replica (#18501)
21d3c89 David Traina committed 1 month ago
perf(score-analytics): route score comparison analytics to the read replica (#18504)
cebca12 David Traina committed 1 month ago
perf(evals): route evaluator job score lookups to the read replica (#18500)
7c404e2 David Traina committed 1 month ago
chore(pricing): document Oct 23 2026 OpenAI retirement batch, Gemini 2.5 access note (#18511)
77f5f9a langfuse-bot committed 1 month ago
fix(ci): let the model price audit list the rows its validator expects (#18514)
a36654d Hassieb Pakzad committed 1 month ago
fix(experiment): update experiment display settings and score column header for clarity (#18323)
c106bb3 marliessophie committed 1 month ago
chore(worker): log project and failure state on queue job failures (#18437)
e82b2ad Niklas Semmler committed 1 month ago
fix(web): keep trace tree rows from painting over the scrollbar (#18453)
226e883 Der Gerät committed 1 month ago
feat(web): ask new users what they will use Langfuse for (#18356)
25edc58 Abdallah Othman committed 1 month ago
fix(mcp): Always require date range in get observations tool (#18485)
0da6dea Ben Bachem committed 1 month ago
feat(billing): send spend alerts for ClickHouse-billed organizations (#17615)
74acb1a Steffen Reuther committed 1 month ago
fix(web): env and plan badges use the design-system Badge (#18481)
508cb5b Trang Nguyen committed 1 month ago
fix(web): keep the previous trace on screen while the next one loads (#18280)
d5f3668 Trang Nguyen committed 1 month ago
fix(web): design-system LabelList with score-chip style labels (#18463)
b6cf2cd Trang Nguyen committed 1 month ago
feat(sessions): Add syntax highlighting to session tool previews (#18472)
81663ae Ben Bachem committed 1 month ago
fix(web): header actions, metrics, scores and +N match across trace, observation and session (#18470)
6f73f27 Trang Nguyen committed 1 month ago