mirror of
https://github.com/Helldez/BigMoeOnEdge.git
synced 2026-10-03 03:25:42 +00:00
docs(readme): decisions in a section of their own, routed arena in the NPU prefill (#209)
ci.yml #20 -Commit
374f562c92
pushed by
vrr
docs(readme): decisions in a section of their own, routed arena in the NPU prefill (#209)
ci.yml #19 -Commit
374f562c92
pushed by
vrr
feat: --decide, choose from a list from one prefill, with no decode (#201)
ci.yml #18 -Commit
1d37057fac
pushed by
vrr
feat: prefill on the NPU, decode on the CPU (--prefill-device), in the release APK (#200)
ci.yml #17 -Commit
10af539a63
pushed by
vrr
feat(moe): Nemotron 3.5 (nemotron_h_moe) and Ornith 1.5 support (#203)
ci.yml #16 -Commit
50d0a38338
pushed by
vrr
feat: --decide, choose from a list from one prefill, with no decode (#201)
ci.yml #15 -Commit
1d37057fac
pushed by
vrr
fix(scripts): bench-report model size and storage probe (du on the raw path) (#194)
ci.yml #12 -Commit
92a1114917
pushed by
vrr
docs(assets): label the Qwen3.8-Flash-Next hero clip like the DeepSeek one
ci.yml #11 -Commit
557c63dda0
pushed by
vrr
feat(app): Qwen3.8-Flash-Next Q2_K in the catalog, the low-dense build (#189)
ci.yml #10 -Commit
f24bd5aecc
pushed by
vrr
docs(bench): tell contributors to check the mode line before reporting a number (#190)
ci.yml #9 -Commit
4c02169fee
pushed by
vrr
fix(cli): make the Windows release exe survive a double click (#183)
ci.yml #8 -Commit
d754dfd6fc
pushed by
vrr
feat(moe): cache-aware expert substitution (--expert-substitute, experimental) and --ppl (#171)
ci.yml #7 -Commit
4334c89616
pushed by
vrr
feat(telemetry): add prefill-phase attribution (#175)
ci.yml #6 -Commit
039b8f1e2f
pushed by
vrr
docs(changelog): Qwen3.8-Flash-Next numbers on the final pin
ci.yml #5 -Commit
2c73fb3a5d
pushed by
vrr
docs(changelog): Qwen3.8-Flash-Next numbers on the final pin
ci.yml #4 -Commit
2c73fb3a5d
pushed by
vrr
ci: prebuilt bmoe-cli for Linux, macOS and Windows on every release (#178)
ci.yml #3 -Commit
f9371b4ddb
pushed by
vrr
fix(telemetry): attribute critical-path stall explicitly (#169)
ci.yml #2 -Commit
7a8abcdda7
pushed by
vrr
feat(moe): hold back dense tensors larger than RAM under every mode
ci.yml #1 -Commit
7b08da1dc7
pushed by
vrr