Commit Graph

  • ac9b3b17af manifest: pin to 20261118-140b443 main inference-admin 2026-11-18 17:58:02 +00:00
  • 140b443980 gateway: refresh EUR->USD prices at boot from the ECB feed (fail-open) inference-admin 2026-11-18 17:57:09 +00:00
  • fe48ca8e38 manifest: pin to 20261118-6fcc116 inference-admin 2026-11-18 17:30:43 +00:00
  • 6fcc116199 scripts: EUR FX guard also covers the route aliases inference-admin 2026-11-18 17:29:38 +00:00
  • c8816b58c1 gateway: convert the route aliases too — members call those, not the prefixed ids inference-admin 2026-11-18 17:29:29 +00:00
  • 718644c363 manifest: pin to 20261118-870b165 inference-admin 2026-11-18 17:26:03 +00:00
  • 870b165321 scripts: check EUR->USD conversions in model_info (FX guard) inference-admin 2026-11-18 17:24:57 +00:00
  • aa9331cff3 gateway: convert GreenPT and Regolo EUR prices to USD at FX 1.16 (corrected) inference-admin 2026-11-18 17:24:39 +00:00
  • 6f6bb73d16 gateway: convert GreenPT and Regolo EUR list prices to USD at FX 1.16 inference-admin 2026-11-18 17:24:23 +00:00
  • 9c81df48b8 gateway: raise Cloudron memory limit 2 GB -> 3 GB inference-admin 2026-11-18 14:13:15 +00:00
  • d1d342d8e6 manifest: pin to 20261117-0f7cf17 inference-admin 2026-11-17 15:19:08 +00:00
  • 0f7cf17cd8 litellm(config): comments match the prices members are actually billed inference-admin 2026-11-17 15:16:05 +00:00
  • 278900684c manifest: pin to 20261117-df4e154 inference-bot 2026-11-17 09:05:26 +00:00
  • df4e154128 audio patch: price per-request models when unit pricing raises (#5) inference-bot 2026-11-17 09:04:28 +00:00
  • 4b6583c3c0 manifest: pin to 20261117-2483715 inference-admin 2026-11-17 08:45:27 +00:00
  • 2483715ef3 litellm: register audio per-request prices at interpreter startup (#5) inference-admin 2026-11-17 08:44:41 +00:00
  • f66bedfb55 manifest: pin to 20261117-0e8517b inference-admin 2026-11-17 08:37:26 +00:00
  • 0e8517b6be litellm: target *_backup.json — LiteLLM's local cost map, not the file name we guessed inference-admin 2026-11-17 08:36:53 +00:00
  • d35ee1760f manifest: revert pin — the audio cost-map build failed, image never existed inference-admin 2026-11-17 08:34:38 +00:00
  • faff547ff4 manifest: pin to 20261117-b81177c inference-admin 2026-11-17 08:33:12 +00:00
  • b81177c3ad litellm: put Tinfoil's per-request audio prices in the shared cost map (#5) inference-admin 2026-11-17 08:32:56 +00:00
  • d96df6fd5a manifest: pin to 20261116-7a167a8 inference-admin 2026-11-16 19:56:17 +00:00
  • 7a167a8354 audio patch: bill speech too — the provider-cost early return pre-empted the tail patch inference-admin 2026-11-16 19:55:43 +00:00
  • a1613401c3 manifest: pin to 20261116-77e38d8 inference-admin 2026-11-16 15:55:57 +00:00
  • 77e38d81e0 gateway: serve the DeepSeek V4.1 Flash 1017 snapshot under the existing names inference-admin 2026-11-16 15:55:23 +00:00
  • d877bc8eee manifest: pin to 20261116-4377611 inference-admin 2026-11-16 15:53:35 +00:00
  • 437761171c manifest: pin to 20261114-84a1918 inference-admin 2026-11-14 18:56:23 +00:00
  • 84a191851f gateway: glm-green -> glm-flash-green, for symmetry with deepseek-flash-green inference-admin 2026-11-14 18:55:53 +00:00
  • 1ec589cf8b manifest: pin to 20261114-9266f9b inference-admin 2026-11-14 18:46:24 +00:00
  • 9266f9bc3c gateway: aliases name the route, not the version (-green / -tee) inference-admin 2026-11-14 18:45:53 +00:00
  • 02ff60bfbf manifest: pin to 20261113-5986ee7 inference-admin 2026-11-14 18:40:47 +00:00
  • 5986ee7987 gateway: stable model aliases members can pin across transitions inference-admin 2026-11-14 18:40:10 +00:00
  • aa1531bb18 gateway: add Regolo Mistral-Small-4-119B and Gemma-4-31B inference-bot 2026-11-13 15:23:47 +00:00
  • e21c24ea48 gateway: fail-closed config validation before boot — unknown top-level key or unresolvable provider prefix refuses the start (ops-docs canonical copy shipped byte-identical) inference-bot 2026-11-11 03:55:18 +00:00
  • 64d7e5afc5 gateway: single TTS warm path — drop the 10-min scheduler task, in-container loop every 15 min inference-bot 2026-11-09 14:33:52 +00:00
  • c43d45c455 upgrade: pin LiteLLM v1.108.3 (re-validated on real 1.108.3 sources) upgrade-v1.108.3 inference-bot 2026-11-09 04:22:30 +00:00
  • 5d38403892 gateway: fix TTS latency + restore audio warm path inference-bot 2026-11-07 20:09:54 +00:00
  • 943701581c gateway: make /v1/audio/transcriptions work bare for voxtral-small-24b inference-bot 2026-11-06 22:40:53 +00:00
  • abaeb255b2 gateway: make tinfoil/nomic-embed-text member-facing; add greenpt/green-embeddings; fix nomic price to $0.016/1M inference-bot 2026-11-06 22:35:58 +00:00
  • 67df968a04 gateway: add tinfoil/voxtral-small-24b, Tinfoil's token-priced STT model inference-bot 2026-11-06 22:33:17 +00:00
  • 54925cae97 gateway: add greenpt/qwen3-embedding-8b (member-facing embeddings) inference-bot 2026-11-06 20:28:05 +00:00
  • 9e25578ee6 upgrade: pin LiteLLM v1.108.0 (candidate for 2026-11-08 window) upgrade-v1.108.0 inference-bot 2026-11-06 03:55:36 +00:00
  • e161db9ec8 docs: state the wire-verified truth about /ui, docs surfaces and CORS (audit M4+L5); manifest contactEmail (L6) inference-bot 2026-11-05 16:16:16 +00:00
  • d6cfe78c77 gateway: record token counts on Metis pass-through spend rows inference-bot 2026-11-03 19:13:09 +00:00
  • da02f8eb32 gateway: price the Metis pass-through route from its reported tokens inference-bot 2026-11-03 18:58:07 +00:00
  • 557999a16f gateway: let member keys reach the Metis pass-through route inference-bot 2026-11-03 18:30:22 +00:00
  • e4e70962ec gateway: expose GreenPT's Metis decision model at /v1/systemone inference-bot 2026-11-03 18:24:49 +00:00
  • e8378d941a gateway: Regolo's 2026-11-01 prices inference-bot 2026-11-01 00:41:42 +00:00
  • 83c543a0ba gateway: expose the eight telemetry headers requested in co-op/support#39 inference-bot 2026-10-31 21:26:03 +00:00
  • 197abce9e6 gateway: honour LITELLM_CORS_ORIGINS from app env; expose cost headers (#38, #39) inference-bot 2026-10-31 17:20:01 +00:00
  • 8ddd023786 gateway: stop forwarding provider response headers to callers inference-bot 2026-10-27 04:00:57 +00:00
  • a3acce2107 gateway: instrument the audio per-request helper (speech still bills $0) Ubuntu 2026-10-26 21:35:02 +00:00
  • a6098741d0 gateway: declare a zero character price for voxtral-tts so speech can bill Ubuntu 2026-10-26 21:29:59 +00:00
  • c03abaa330 gateway: put audio pricing on the shared cost-map key (no SpendLogs row at all) Ubuntu 2026-10-26 21:24:42 +00:00
  • facf01d578 gateway: deploy config.yaml changes (the volume was shadowing the image) Ubuntu 2026-10-26 21:17:41 +00:00
  • 425d738625 gateway: bill Tinfoil's per-request charges on the audio paths fix/audio-request-cost Ubuntu 2026-10-26 20:45:06 +00:00
  • c877b61ef3 voxtral-tts: bill speech at the model's published rate ($16/1M characters) fix/voxtral-tts-pricing Ubuntu 2026-10-26 20:17:06 +00:00
  • f75851d6cf start.sh: kill stale tinfoil-proxy before bind + fail hard if the proxy never comes up Inferencebot 2026-10-22 08:25:11 -06:00
  • 79f0447089 gateway: build-time patch keeping image-gen pricing on the shared cost-map key (upstream #10023/#12471) so image spend is tracked Inferencebot 2026-10-19 09:26:45 -06:00
  • 9d95527e93 Remove PublicAI provider (billing never operational; Regolo replaces it for Apertus) Inferencebot 2026-10-16 10:46:04 -06:00
  • ad34da72a4 Add Regolo provider: apertus-70b, qwen3.5-122b (thinking off), qwen-image + Apertus fallback chain Inferencebot 2026-10-16 10:10:45 -06:00
  • d082957167 README: document the audio keep-alive scheduled task Inferencebot 2026-10-12 19:33:18 -06:00
  • 0d73c80562 keepalive: read master key from /app/data/.master_key (scheduler runs outside litellm's env) Inferencebot 2026-10-12 19:20:24 -06:00
  • acceb43995 Move audio keep-alive into the gateway container (Cloudron scheduler addon, every 10m) Inferencebot 2026-10-12 19:18:46 -06:00
  • 3397d07ec1 Replace tinfoil/glm-5-3-flash (deprecated 2026-10-09) with tinfoil/glm-5.3 flagship Inferencebot 2026-10-07 22:07:40 -06:00
  • cc8f7d237a Add tinfoil/nomic-embed-text embedding model (admin-team scoped, for co-op docs RAG) Inferencebot 2026-10-07 09:44:12 -06:00
  • 5389d9acfc credit-pack hint: patch budget-exceeded message at BUILD time (runtime rootfs is read-only; boot-time patch crashed startup) Inferencebot 2026-10-06 14:24:55 -06:00
  • 91aa0e0b1d start.sh: append credit-pack hint to LiteLLM's hardcoded budget-exceeded messages (team + key) Inferencebot 2026-10-06 14:04:22 -06:00
  • 1f2c330762 greenpt/glm-5.3: add cache-read price (0.275/1M per GreenPT pricing API) — member PR follow-up Inferencebot 2026-10-05 08:40:13 -06:00
  • a0ae99329f Add greenpt/deepseek-v4.1-flash as default (renewable, not Experimental); fallback → tinfoil deepseek Inferencebot 2026-10-02 22:15:44 -06:00
  • 516167aaf9 Remove green-l/green-r models (GreenPT marks them not recommended) Inferencebot 2026-10-02 22:04:47 -06:00
  • b6e8542146 config: remove duplicate broken audio-model entries (audio_transcription//audio_speech/ prefixes); keep the correct openai/ routing Inferencebot 2026-10-02 15:39:07 -06:00
  • b9af62a614 README: current base image v1.103.2 Inferencebot 2026-10-02 15:32:49 -06:00
  • a2864ce34a Upgrade to LiteLLM v1.103.2 (schema pre-migrated live; CVE fix line + streaming-usage bugfix window) Inferencebot 2026-10-02 15:25:27 -06:00
  • 23f73cf6ae README: current architecture (3 providers, portal key injection, hardening), remove MVP model list (point to co-op/docs as source of truth), all provider env vars, member-key verification kept Inferencebot 2026-10-02 01:08:34 -06:00
  • e8c7a55933 Add upgrade runbook + bake memory limit and LITELLM_MIGRATION_DIR into packaging fix/cve-upgrade-v1.84.0-repo-sync inference-co-op-bot 2026-10-02 00:52:50 -06:00
  • fd34e33508 Sync packaging to live v1.84.0 deployment (CVE-2026-35029/59822 fix line) inference-bot 2026-10-01 23:32:43 -06:00
  • 8f42be988e Add Tinfoil audio models (whisper STT, voxtral-tts) for voice mode + API use inference-bot 2026-09-27 11:10:51 -06:00
  • 5ea2256776 Add greenpt/glm-5.3 (full flagship, $1.10/$4.40 per 1M) inference-bot 2026-09-26 16:52:21 -06:00
  • 057507d67c Add PublicAI Apertus models (publicai/apertus-v1.5-8b, -70b) inference-bot 2026-09-25 15:56:08 -06:00
  • 2369c3e848 Add greenpt/glm-5.3-flash (/usr/bin/bash.11//usr/bin/bash.44 per 1M) inference-bot 2026-09-24 10:43:14 -06:00
  • f86e005062 Rename Tinfoil models to tinfoil/* provider prefix inference-bot 2026-09-24 08:16:16 -06:00
  • 67192de232 Document pricing sync with co-op/docs/models.md inference-bot 2026-09-23 21:08:33 -06:00
  • 09ee36014c Add GreenPT models (greenpt/green-r, greenpt/green-l) via provider/model naming convention inference-bot 2026-09-23 20:10:18 -06:00
  • d1f57c45d5 Retry connection errors but never re-bill timeouts inference-bot 2026-09-22 18:18:07 -06:00
  • 576885ed21 Fix Tinfoil spend gap: raise timeout 30->180s, num_retries 2->0 inference-bot 2026-09-22 18:05:27 -06:00
  • 9cbbc33bc7 Fix CORS patch to target site-packages (the copy the running CLI imports) inference-bot 2026-09-14 18:01:13 -06:00
  • 0a797ba50e Patch hardcoded origins=[*] to honour LITELLM_CORS_ALLOWED_ORIGINS (v1.74.0 ignores the env var) inference-bot 2026-09-14 17:55:08 -06:00
  • dae9bf4f31 Lock CORS to chat origin (was wildcard * with allow-credentials) inference-bot 2026-09-14 17:51:38 -06:00
  • fb921b6e09 Disable admin UI (/ui) and API docs (Swagger/ReDoc/openapi) on the public gateway inference-bot 2026-09-14 16:21:34 -06:00
  • 700154c3d4 Add custom per-token pricing (model_info) so spend and budget enforcement work inference-bot 2026-09-14 14:28:13 -06:00
  • da50b0438f Replace deprecated deepseek-v4-flash with deepseek-v4-1-flash (default + fallback) inference-bot 2026-09-14 14:06:23 -06:00
  • 878ab8fa8d Add GLM-5.3 Flash model inference-bot 2026-09-09 14:18:14 -06:00
  • 8a54e75d4e Download tinfoil-proxy at runtime (build sandbox has no GitHub access) inference-bot 2026-09-09 13:23:53 -06:00
  • 26b0edc6a8 Hardcode tinfoil-proxy version in Dockerfile inference-bot 2026-09-09 13:22:21 -06:00
  • e05fec0ca8 Switch to Tinfoil (TEE): add tinfoil-proxy sidecar for EHBP encryption, point LiteLLM at local proxy inference-bot 2026-09-09 13:21:01 -06:00
  • 2ce5c66b9b Switch to Ollama Cloud (OpenAI-compatible /v1 endpoint) for plumbing smoke test inference-bot 2026-09-04 14:15:49 -06:00
  • 41068a6264 Fix Dockerfile for Cloudron: run as root (base image default), override ENTRYPOINT, fix .dockerignore inference-bot 2026-09-04 10:29:15 -06:00
  • b9b4620e52 Initial commit: LiteLLM Cloudron app package for Inference Cooperative inference-bot 2026-09-03 21:55:54 -06:00