README: gold6 numbers (mean retention 95.1% -> 96.3%)

README.md

@@ -22,7 +22,7 @@ tags:
 
 | | Mean retention, 12 benchmarks |
 | --- | ---: |
-| **Mach-1 Small** | **95.1%** |
+| **Mach-1 Small** | **96.3%** |
 | Ternary Bonsai 27B (PrismML) | 93.6% |
 | Gemma 4 Q2_K_XL (Unsloth) | 85.6% |
 
@@ -30,36 +30,36 @@ tags:
 
 | Benchmark | Mach-1 Small | Ternary Bonsai 27B | Gemma 4 Q2_K_XL |
 | --- | ---: | ---: | ---: |
-| AIME26 | **93.5%** | 92.7% | 67.7% |
-| MATH-500 | **98.2%** | **98.2%** | 95.6% |
-| AIME25 | **92.5%** | 91.7% | 67.2% |
-| GSM8K | 99.0% | **100.2%** | 97.3% |
-| MBPP+ | 93.9% | **98.4%** | 92.2% |
-| HumanEval+ | 91.7% | **98.7%** | 94.1% |
-| MMLU-Redux | 96.0% | 94.0% | **96.9%** |
-| IFEval | **96.5%** | 89.8% | 95.5% |
-| MuSR | **97.6%** | 91.6% | 91.1% |
+| AIME26 | **95.8%** | 92.7% | 67.7% |
+| MATH-500 | **99.2%** | 98.2% | 95.6% |
+| AIME25 | **93.9%** | 91.7% | 67.2% |
+| GSM8K | 99.8% | **100.2%** | 97.3% |
+| MBPP+ | 98.1% | **98.4%** | 92.2% |
+| HumanEval+ | 96.5% | **98.7%** | 94.1% |
+| MMLU-Redux | **97.0%** | 94.0% | 96.9% |
+| IFEval | 94.8% | 89.8% | **95.5%** |
+| MuSR | **100.4%** | 91.6% | 91.1% |
 | BFCL-v3 | 96.3% | **98.9%** | 95.7% |
-| τ²-bench | **95.5%** | 91.2% | 73.1% |
-| IFBench | **91.1%** | 77.7% | 61.3% |
-| **Mean** | **95.1%** | **93.6%** | **85.6%** |
+| τ²-bench | **94.1%** | 91.2% | 73.1% |
+| IFBench | **89.8%** | 77.7% | 61.3% |
+| **Mean** | **96.3%** | 93.6% | 85.6% |
 
 Mach-1 Small's own scores and teacher scores:
 
 | Benchmark | Score | Teacher (Qwen3.6-35B-A3B BF16) | Retention |
 | --- | ---: | ---: | ---: |
-| AIME26 | 84.17 | 90.00 | 93.5% |
-| MATH-500 | 96.80 | 98.60 | 98.2% |
-| AIME25 | 81.67 | 88.33 | 92.5% |
-| GSM8K | 95.22 | 96.21 | 99.0% |
-| MBPP+ | 90.21 | 96.03 | 93.9% |
-| HumanEval+ | 87.20 | 95.12 | 91.7% |
-| MMLU-Redux | 88.93 | 92.68 | 96.0% |
-| IFEval | 85.90 | 89.05 | 96.5% |
-| MuSR | 65.08 | 66.66 | 97.6% |
-| BFCL-v3 | 72.21 | 74.98 | 96.3% |
-| τ²-bench | 75.90 | 79.51 | 95.5% |
-| IFBench | 59.18 | 64.97 | 91.1% |
+| AIME26 | 86.25 | 90.00 | 95.8% |
+| MATH-500 | 97.80 | 98.60 | 99.2% |
+| AIME25 | 82.92 | 88.33 | 93.9% |
+| GSM8K | 95.98 | 96.21 | 99.8% |
+| MBPP+ | 94.18 | 96.03 | 98.1% |
+| HumanEval+ | 91.77 | 95.12 | 96.5% |
+| MMLU-Redux | 89.88 | 92.68 | 97.0% |
+| IFEval | 84.42 | 89.05 | 94.8% |
+| MuSR | 66.93 | 66.66 | 100.4% |
+| BFCL-v3 | 72.19 | 74.98 | 96.3% |
+| τ²-bench | 74.82 | 79.51 | 94.1% |
+| IFBench | 58.33 | 64.97 | 89.8% |
 
 ### Speed