Update halo-models.md

nod-ai · Dec 2, 2024 · 6ff7b3e · 6ff7b3e
1 parent d907df4
commit 6ff7b3e
Showing 1 changed file with 3 additions and 3 deletions.
diff --git a/halo-models.md b/halo-models.md
@@ -32,10 +32,10 @@ See latest [CI/Nightly Test Report](https://nod-ai.github.io/shark-ai/). Use [No
 (Model is assumed to be llama3.1 in the following table, e.g. "8B FP8" means "llama3.1 8B FP8 model")
 |Item                          | Last Week (Nov 25-27) | Current Week (Dec 2-6) |
 |------------------------------|-----------------------|--------------------------|
-| Sharktank Modeling           | <br> - @Boian Add CLIP encoder to sharktank (ETA: 11/27) <br> -@Dan fix numeric fp8 issue (ETA: 11/26) |
+| Sharktank Modeling           | <br> - @Boian Add CLIP encoder to sharktank (ETA: 11/27) <br> -@Dan fix numeric fp8 issue (ETA: 11/26) | - @Boian CLIP encoder (ETA: 12/5) <br> - @Rob CI llama regression tests (ETA 12/3) <br> -@Ian 
 | IREE codegeneration          |- @kunvar support for non deocmposed decode (ETA: 11/27) <br> - @stan: FP8 attention (ETA: 11/27) | 
-| Serving | <br> - @stephen / @xida implement Radix Attention in shortfin (ETA: 12/6) <br> - @egarvey wire up Flux.1 in shortfin using ONNX model (ETA:11/27) |
-| Test Automation              |<br>- @Avi Work with codegen folks to get 405B FP16 fixed and tested (ETA: 11/18) 
+| Serving | <br> - @stephen / @xida implement Radix Attention in shortfin (ETA: 12/6) <br> - @egarvey wire up Flux.1 in shortfin using ONNX model (ETA:11/27) | - @eagarvey finish Flux pipeline for image generation (ETA: 12/2) | - @Stephen Debug CI flakiness (ETA: 12/2) <br> - @Xida landing PR's for attention changes (ETA: 12/3)
+| Test Automation              |<br>- @Avi Work with codegen folks to get 405B FP16 fixed and tested (ETA: 11/18) | - @Avi benchmarking dashboard (ETA: 12/3) <br> - @Archana shortfin regression tests (ETA: 12/3)
 | Performance Tuning           | |