Keep MLX draft comparisons and highlight only column-best speeds

README.md

@@ -27,29 +27,15 @@ This checkpoint is the **DFlash DPACE branch** adapted to:
 
 ## Results
 
-### A100 / SGLang
-
-| Draft | Overall tok/s |
-| --- | ---: |
-| Target only | 175.60 |
-| Official DSpark | **354.69** |
-| Draftfit DSpark | **347.94** |
-| Draftfit DFlash CE | 294.77 |
-| **Draftfit DFlash DPACE (Mine)** | **304.85** |
-| DFlash2 | not measured |
-
 ### MLX / M5 Pro
 
 | Draft | Overall tok/s |
 | --- | ---: |
-| Target only | 50.98 |
 | Official DSpark | 105.38 |
 | Draftfit DSpark | **116.15** |
-| DFlash CE | 89.61 |
-| **Draftfit DFlash DPACE (Mine)** | **89.97** |
-| DFlash2 | 80.75 |
-
-A100 and MLX use different measurement methods, so their absolute numbers should **not** be compared directly.
+| Draftfit DFlash CE | 89.61 |
+| Draftfit DFlash DPACE | 89.97 |
+| Draftfit DFlash2 | 80.75 |
 
 Validation set: **53 held-out conversations**
 
@@ -62,12 +48,11 @@ Validation set: **53 held-out conversations**
 
 | Draft | PerfectBlend | OpenCode | APIGen-MT | Nemotron Agentic |
 | --- | ---: | ---: | ---: | ---: |
-| Target only | 50.76 | 51.04 | 51.47 | 51.81 |
 | Official DSpark | 107.83 | 112.19 | 107.79 | 85.86 |
 | Draftfit DSpark | **118.29** | **122.28** | **116.04** | **91.99** |
-| DFlash CE | 89.02 | 99.18 | 92.49 | 79.31 |
-| **Draftfit DFlash DPACE (Mine)** | **89.84** | **103.20** | **91.62** | **76.97** |
-| DFlash2 | 78.37 | 92.76 | 87.49 | 70.41 |
+| Draftfit DFlash CE | 89.02 | 99.18 | 92.49 | 79.31 |
+| Draftfit DFlash DPACE | 89.84 | 103.20 | 91.62 | 76.97 |
+| Draftfit DFlash2 | 78.37 | 92.76 | 87.49 | 70.41 |
 
 Full prompt-free summary: [`evaluation-v4.json`](evaluation-v4.json)
 
@@ -75,14 +60,14 @@ Full prompt-free summary: [`evaluation-v4.json`](evaluation-v4.json)
 
 | Draft | Best use |
 | --- | --- |
-| **DSpark** | Best first choice when adapting an existing DSpark draft to a new target |
-| **DFlash CE** | Simple block-parallel DFlash baseline |
-| **DFlash DPACE** | Test position-weighted training objectives against CE |
-| **DFlash2** | Research richer candidate ranking / grouped-convolution drafts |
+| DSpark | Best first choice when adapting an existing DSpark draft to a new target |
+| DFlash CE | Simple block-parallel DFlash baseline |
+| DFlash DPACE | Test position-weighted training objectives against CE |
+| DFlash2 | Research richer candidate ranking / grouped-convolution drafts |
 
 For this MiniCPM5-SFT experiment, **DSpark is the strongest practical baseline**.
 
-Draftfit DSpark reached **347.94 tok/s vs 354.69 tok/s** for the official DSpark on A100, while reaching **116.15 vs 105.38 tok/s** in the tested MLX setup.
+Draftfit DSpark reached **116.15 tok/s vs 105.38 tok/s** for the official DSpark in the tested MLX setup.
 
 These measurements do **not** establish a general speed advantage.
 
@@ -93,12 +78,7 @@ These measurements do **not** establish a general speed advantage.
 | `main` | DFlash DPACE, primary release |
 | `v5` | Additional 1,000-step continual-training stage on a fresh split |
 
-Best observed DSpark A100 result:
-
-- Draftfit DSpark [`v3-best`](https://huggingface.co/jayyun98/MiniCPM5-2B-SFT-Draftfit-DSpark/tree/v3-best): **360.26 tok/s**
-- Matching official DSpark: **361.36 tok/s**
-
-Different revisions used different held-out sets, so these numbers should not be used to rank revisions directly.
+Different revisions used different held-out sets, so their measurements should not be used to rank revisions directly.
 
 ## Validation notes
 
@@ -116,7 +96,7 @@ Training mixture per stage:
 Important limitations:
 
 - DSpark/DFlash serving correctness still requires exact token/state and acceptance-length validation.
-- DFlash2 was tested on MLX but not on A100 with SGLang 0.5.16.
+- DFlash2 was tested on MLX but not with SGLang 0.5.16.
 - A successfully loaded draft checkpoint does **not** prove correct speculative decoding.
 - Reported speed is specific to the tested target, hardware, verifier settings and prompt set.