-
Notifications
You must be signed in to change notification settings - Fork 140
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
docs: define a currency policy for docs/dev/ design notes (several are stale/misleading)
documentationImprovements or additions to documentationImprovements or additions to documentationneeds-designProblem is clear; implementation approach needs design discussion before work startsProblem is clear; implementation approach needs design discussion before work startsp2Medium/low: minor bugs, niche features, polish, docs, tests, cleanup. Scoped, lower urgency.Medium/low: minor bugs, niche features, polish, docs, tests, cleanup. Scoped, lower urgency.Status: Open.#1482 In generative-computing/mellea;- Status: Open.#1479 In generative-computing/mellea;
Generative stub prompt template underperforms on smaller models
area/componentComponent type system, CBlock, @generative decorator, prompt templates, RichDocumentComponent type system, CBlock, @generative decorator, prompt templates, RichDocumentenhancementNew feature or requestNew feature or requestneeds-designProblem is clear; implementation approach needs design discussion before work startsProblem is clear; implementation approach needs design discussion before work startsp1High: important bugs (workaround exists) or high-value core features. Do soon, not on fire.High: important bugs (workaround exists) or high-value core features. Do soon, not on fire.Status: Open.- Status: Open.#1475 In generative-computing/mellea;
Clarify Switch adapter docs and show adapters working together
area/adapter-functionsGranite adapter functions: framework and adaptiers including RAG, Guardian, CoreGranite adapter functions: framework and adaptiers including RAG, Guardian, CoredocumentationImprovements or additions to documentationImprovements or additions to documentationp1High: important bugs (workaround exists) or high-value core features. Do soon, not on fire.High: important bugs (workaround exists) or high-value core features. Do soon, not on fire.Status: Open.Guidance on when to use Python validators vs. LLM-as-a-judge vs. aLoRA
area/requirementRequirement base class, validation, repair loopsRequirement base class, validation, repair loopsdocumentationImprovements or additions to documentationImprovements or additions to documentationp1High: important bugs (workaround exists) or high-value core features. Do soon, not on fire.High: important bugs (workaround exists) or high-value core features. Do soon, not on fire.Status: Open.Make backends extensible without editing core stdlib internals
area/backendsProvider-specific work: Ollama, HF, LiteLLM, OpenAI, Bedrock, vLLMProvider-specific work: Ollama, HF, LiteLLM, OpenAI, Bedrock, vLLMhelp wantedExtra attention is neededExtra attention is neededneeds-designProblem is clear; implementation approach needs design discussion before work startsProblem is clear; implementation approach needs design discussion before work startsStatus: Open.#1467 In generative-computing/mellea;feat(telemetry): emit adapter-function spans from a tracing plugin (Epic #929 Phase 2)
area/adapter-functionsGranite adapter functions: framework and adaptiers including RAG, Guardian, CoreGranite adapter functions: framework and adaptiers including RAG, Guardian, Corearea/telemetryOTel spans, metrics, tracing, semconvOTel spans, metrics, tracing, semconvenhancementNew feature or requestNew feature or requestp1High: important bugs (workaround exists) or high-value core features. Do soon, not on fire.High: important bugs (workaround exists) or high-value core features. Do soon, not on fire.Status: Open.#1466 In generative-computing/mellea;refactor(backends): route intrinsic generation through adapter_scope (Epic #929 Phase 2)
area/adapter-functionsGranite adapter functions: framework and adaptiers including RAG, Guardian, CoreGranite adapter functions: framework and adaptiers including RAG, Guardian, Corearea/aloraLoRA/aLora adapter trainingLoRA/aLora adapter trainingarea/backendsProvider-specific work: Ollama, HF, LiteLLM, OpenAI, Bedrock, vLLMProvider-specific work: Ollama, HF, LiteLLM, OpenAI, Bedrock, vLLMp1High: important bugs (workaround exists) or high-value core features. Do soon, not on fire.High: important bugs (workaround exists) or high-value core features. Do soon, not on fire.Status: Open.#1465 In generative-computing/mellea;docs(telemetry): document the hook/plugin span-production rule and enforce it (Epic #929)
area/adapter-functionsGranite adapter functions: framework and adaptiers including RAG, Guardian, CoreGranite adapter functions: framework and adaptiers including RAG, Guardian, Corearea/telemetryOTel spans, metrics, tracing, semconvOTel spans, metrics, tracing, semconvdocumentationImprovements or additions to documentationImprovements or additions to documentationp1High: important bugs (workaround exists) or high-value core features. Do soon, not on fire.High: important bugs (workaround exists) or high-value core features. Do soon, not on fire.Status: Open.#1464 In generative-computing/mellea;litellm resolves to different minors across the Python 3.12–3.14 boundary
choreHousekeeping: renames, comment fixes, dependency bumps, repo hygieneHousekeeping: renames, comment fixes, dependency bumps, repo hygieneStatus: Open.#1456 In generative-computing/mellea;feat(backends): add default_extra_body to OpenAIBackend for vLLM/vendor-specific per-request fields
enhancementNew feature or requestNew feature or requestStatus: Open.#1453 In generative-computing/mellea;