-
Upload reports/chart-grpo-gsm8k.png with huggingface_hub
-
Upload reports/fp8-rl-grpo.md with huggingface_hub
-
Upload README.md with huggingface_hub
-
Upload reports/chart-nvfp4-vs-q6.png with huggingface_hub
-
Upload reports/nvfp4-vs-q6-qwen3-6-27b.md with huggingface_hub
-
Upload reports/chart-lfm2-vl-extract.png with huggingface_hub
-
Upload reports/lfm2-5-vl-1-6b-extract.md with huggingface_hub
-
Add Nemotron-3-Nano-30B-A3B to think-off leaderboard + note format-agnostic HumanEval fix
-
Add Nemotron-3-Nano speed-vs-quality chart
-
Add Nemotron-3-Nano-30B-A3B benchmark report
-
Update dataset card: split quality table by think mode + corrected HumanEval (all models) + add Gemma 4 12B
-
Correct HumanEval (harness bug) + publish Gemma 4 12B + quality leaderboard card
-
REAP vs parent matched comparison
-
REAP vs parent matched comparison
-
REAP vs parent matched comparison
-
Add gpt-oss-120B + REAP rows; drop pre-fix humaneval rows
-
Add gpt-oss-120B + REAP; withhold pre-fix HumanEval (reasoning-aware harness)
-
quality-vs-active-params chart
-
gpt-oss-120b MXFP4 5090 report
-
update nvidia-nemotron-cascade-2-30b-a3b-q4-k-m report with extended context data
-
update qwen3-coder-next-ud-q2-k-xl report with extended context data
-
update qwen3-6-35b-a3b-ud-q4-k-m report with extended context data
-
update qwen3-6-27b-q6-k report with extended context data
-
add extended context speed data (32K, 64K, 128K)
-
add Nemotron-Cascade-2 report
-
add Gemma 4 31B report
-
add Qwen3-Coder-Next report
-
complete 5-model quality leaderboard
-
add Qwen3-Coder-Next, Gemma 4 31B, Nemotron-Cascade-2 results
-
add Qwen3.6-35B-A3B full benchmark report
-
add Qwen3.6-35B-A3B to quality table
-
add Qwen3.6-35B-A3B speed + quality results
-
add Qwen3.6-27B Q6_K full benchmark report
-
add Qwen3.6-27B Q6_K to quality table
-
add Qwen3.6-27B Q6_K speed + quality results
-
Upload README.md with huggingface_hub
-
Upload reports/gpt-oss-20b-q4-k-m.md with huggingface_hub
-
Upload benchmarks.csv with huggingface_hub
-
Upload folder using huggingface_hub
-
initial commit
