-
Notifications
You must be signed in to change notification settings - Fork 41
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
fix(server): preemption victim selection falls back to HashMap order when sequences tie
area:architectureArchitecture and code structure changesArchitecture and code structure changesarea:inferenceGeneration, sampling, decoding (incl. speculative, DRY)Generation, sampling, decoding (incl. speculative, DRY)priority:mediumMedium priorityMedium prioritystatus:readyReady to be worked onReady to be worked ontype:bugBug fixes, error corrections, or issue resolutionsBug fixes, error corrections, or issue resolutionsStatus: Open.#1293 In lablup/mlxcel;refactor(server): give the four LRU eviction min_by_key calls a total order
area:architectureArchitecture and code structure changesArchitecture and code structure changesarea:inferenceGeneration, sampling, decoding (incl. speculative, DRY)Generation, sampling, decoding (incl. speculative, DRY)priority:lowLow priorityLow prioritystatus:readyReady to be worked onReady to be worked ontype:refactorCode restructuring without changing functionalityCode restructuring without changing functionalityStatus: Open.#1291 In lablup/mlxcel;perf(metal): an order-preserving streamed qmv would give MTP the fast kernel and byte-identity together
area:inferenceGeneration, sampling, decoding (incl. speculative, DRY)Generation, sampling, decoding (incl. speculative, DRY)platform:macosmacOS (Apple Silicon) specificmacOS (Apple Silicon) specificpriority:mediumMedium priorityMedium prioritystatus:readyReady to be worked onReady to be worked ontype:performancePerformance improvementsPerformance improvementsStatus: Open.#1289 In lablup/mlxcel;fix(speculative): the Gemma 4 31B pairing's exactness probe fails both kernels on M3 Ultra, vetoing the burst #1217 just enabled
area:inferenceGeneration, sampling, decoding (incl. speculative, DRY)Generation, sampling, decoding (incl. speculative, DRY)platform:macosmacOS (Apple Silicon) specificmacOS (Apple Silicon) specificpriority:highHigh priorityHigh prioritystatus:readyReady to be worked onReady to be worked ontype:bugBug fixes, error corrections, or issue resolutionsBug fixes, error corrections, or issue resolutionsStatus: Open.#1279 In lablup/mlxcel;feat: Bucket the OpenXLA context capacity so text and image requests do not share one graph shape
area:inferenceGeneration, sampling, decoding (incl. speculative, DRY)Generation, sampling, decoding (incl. speculative, DRY)priority:mediumMedium priorityMedium prioritystatus:in-progressCurrently being worked onCurrently being worked ontype:enhancementNew features, capabilities, or significant additionsNew features, capabilities, or significant additionsStatus: Open.#1271 In lablup/mlxcel;[nightly-verify] main is red
priority:highHigh priorityHigh prioritystatus:readyReady to be worked onReady to be worked ontype:bugBug fixes, error corrections, or issue resolutionsBug fixes, error corrections, or issue resolutionsStatus: Open.#1256 In lablup/mlxcel;fix(server): bound the responses and conversation stores by bytes, not just entry count
area:inferenceGeneration, sampling, decoding (incl. speculative, DRY)Generation, sampling, decoding (incl. speculative, DRY)priority:mediumMedium priorityMedium prioritystatus:readyReady to be worked onReady to be worked ontype:bugBug fixes, error corrections, or issue resolutionsBug fixes, error corrections, or issue resolutionsStatus: Open.#1248 In lablup/mlxcel;docs: fix the docs/README.md index intro and entry separator inconsistency
area:docsUser and developer documentationUser and developer documentationgood first issueGood for newcomersGood for newcomerspriority:lowLow priorityLow prioritystatus:readyReady to be worked onReady to be worked ontype:docsDocumentation improvements or additionsDocumentation improvements or additionsStatus: Open.#1247 In lablup/mlxcel;refactor(cli): deduplicate eleven identical "Failed to load image" closures into one open_image helper
area:cliCommand-line interface / CLI flagsCommand-line interface / CLI flagsgood first issueGood for newcomersGood for newcomerspriority:lowLow priorityLow prioritystatus:readyReady to be worked onReady to be worked ontype:refactorCode restructuring without changing functionalityCode restructuring without changing functionalityStatus: Open.#1245 In lablup/mlxcel;refactor(server): split the 8,130-line BatchScheduler into a scheduler/ directory module
area:architectureArchitecture and code structure changesArchitecture and code structure changesarea:inferenceGeneration, sampling, decoding (incl. speculative, DRY)Generation, sampling, decoding (incl. speculative, DRY)priority:mediumMedium priorityMedium prioritystatus:readyReady to be worked onReady to be worked ontype:refactorCode restructuring without changing functionalityCode restructuring without changing functionalityStatus: Open.#1243 In lablup/mlxcel;fix(bench): speculative_bench panics on a user-supplied --target path with no file name
area:benchmarkBenchmark harness and performance measurement (bench_*.sh, /update-benchmarks)Benchmark harness and performance measurement (bench_*.sh, /update-benchmarks)area:cliCommand-line interface / CLI flagsCommand-line interface / CLI flagsgood first issueGood for newcomersGood for newcomerspriority:lowLow priorityLow prioritystatus:readyReady to be worked onReady to be worked ontype:bugBug fixes, error corrections, or issue resolutionsBug fixes, error corrections, or issue resolutionsStatus: Open.#1242 In lablup/mlxcel;chore(cli): unify error-message capitalization across CLI commands
area:cliCommand-line interface / CLI flagsCommand-line interface / CLI flagsgood first issueGood for newcomersGood for newcomerspriority:lowLow priorityLow prioritystatus:readyReady to be worked onReady to be worked ontype:choreMaintenance tasks (build, CI, etc.)Maintenance tasks (build, CI, etc.)Status: Open.#1236 In lablup/mlxcel;