Release v1.21.0-beta.4

What's Changed

• fix(proxy): accept large websocket response.create and fail fast with 400 by @Soju06 in #1236
• perf(request-logs): skip-scan unfiltered filter-option facets by @Soju06 in #1239
• perf(proxy): cut pre-upstream DB round trips on interactive turns by @Soju06 in #1241
• perf(proxy): make SSE event framing linear instead of quadratic by @Soju06 in #1242
• perf(clients): keep upstream connections and DNS across interactive turns by @Soju06 in #1245
• perf(dashboard): compress responses, cache assets, lazy-load charts by @Soju06 in #1243
• perf(usage): aggregate the summary window in SQL, drop dead log refresh by @Soju06 in #1244
• perf(accounts): serve request-usage summaries from persistent rollup by @Soju06 in #1238
• feat(retention): opt-in pruning for request logs and usage history by @Soju06 in #1240
• fix(db): serialize startup migrations across replicas and report schema-ahead correctly by @Soju06 in #1256
• fix(quota-planner): make warmup budget and dedup claims atomic across replicas by @Soju06 in #1260
• chore(openspec): bulk-sync delta specs and archive all 152 completed changes by @Soju06 in #1252
• fix(balancer): persist rate-limit cooldowns so peer replicas honor them by @Soju06 in #1261
• fix(deploy): repair broken multi-replica deployment artifacts by @Soju06 in #1263
• fix(proxy): fence bridge lease writes, evict fenced-out replicas, purge orphaned ring state by @Soju06 in #1259
• fix(caching): extend cross-replica cache invalidation bus to routing, selection, and settings caches by @Soju06 in #1255
• feat(replica-ops): document replica topology contract and land startup guardrails by @Soju06 in #1264
• feat(proxy): partition account concurrency caps across live replicas by @Soju06 in #1258
• fix(reset-credits): make redemption replica-safe with durable ledger and claim by @Soju06 in #1257
• fix(proxy): queue bridged requests through response-create gate contention by @Soju06 in #1266
• feat(balancer): adapt selection and refresh to absent short-window quotas by @Soju06 in #1267
• feat(planner): derive phase planning from observed short windows by @Soju06 in #1268
• perf(proxy): parse each SSE event once per layer, skip no-op reserialization by @Soju06 in #1270
• perf(proxy): detach stream-end persistence from the response path by @Soju06 in #1269
• chore(openspec): archive the ten merged performance changes by @Soju06 in #1276
• perf(request-logs): cache the listing total per filter signature by @Soju06 in #1277
• perf(frontend): code-split dashboard routes by @Soju06 in #1278
• perf(frontend): self-host JetBrains Mono, drop Google Fonts by @Soju06 in #1279
• feat(models): replicate leader-refreshed model registry via persisted snapshots by @Soju06 in #1262
• docs(openspec): archive adaptive-window and bridge-gate changes by @Soju06 in #1271
• fix(automation): fall back to github.token when label sync hits rate limits by @Soju06 in #1272
• feat(usage): ingest live rate-limit snapshots from proxied traffic by @Soju06 in #1273
• docs(openspec): archive the live-rate-limit-ingestion change by @Soju06 in #1300
• feat(dashboard): hide expired primary windows instead of freezing them by @Soju06 in #1301
• feat(ci): mint a dedicated GitHub App token for codex label sync by @Soju06 in #1302
• chore(ci): bump dorny/paths-filter from 4.0.1 to 4.0.2 by @dependabot[bot] in #1288
• chore(ci): bump github/codeql-action/upload-sarif from 4.36.2 to 4.37.0 by @dependabot[bot] in #1289
• chore(ci): bump softprops/action-gh-release from 3.0.1 to 3.0.2 by @dependabot[bot] in #1290
• chore(docker): bump astral-sh/uv from 0.11.27 to 0.11.28 by @dependabot[bot] in #1291
• chore(ci): bump actions/stale from 10.3.0 to 10.4.0 by @dependabot[bot] in #1285
• chore(docker): bump pgautoupgrade/pgautoupgrade from 18-alpine to 18-alpine by @dependabot[bot] in #1286
• chore(ci): bump astral-sh/setup-uv from 8.3.0 to 8.3.2 by @dependabot[bot] in #1287
• chore(deps): bump the frontend-minor-patch group in /frontend with 10 updates by @dependabot[bot] in #1292
• docs(openspec): archive absent-window-display and label-sync-app-token by @Soju06 in #1303
• chore(deps): bump the python-minor-patch group with 5 updates by @dependabot[bot] in #1294
• chore(deps): bump typescript from 6.0.3 to 7.0.2 in /frontend by @dependabot[bot] in #1293
• fix(scheduling): harden scheduler leader election for multi-replica safety by @Soju06 in #1253
• chore: release v1.21.0-beta.4 by @Soju06 in #1237

Full Changelog: v1.21.0-beta.3...v1.21.0-beta.4

Install / Run

uvx --from "codex-lb==1.21.0b4" codex-lb
docker pull ghcr.io/Soju06/codex-lb:1.21.0-beta.4
helm install codex-lb oci://ghcr.io/soju06/charts/codex-lb --version 1.21.0-beta.4 --devel