RedHatAI/Kimi-K2.6-NVFP4

Update BFCL/SWE-Bench eval table + add recovery-vs-base-model comparison for reasoning benchmarks

Update SWE-Bench Lite eval results (official harness resolved-rate + submission-rate, vLLM 0.24.0)

Update BFCL v4 eval results (fresh run on vLLM 0.24.0, corrupted categories re-collected)

Align SWE row style with RedHatAI benchmark cards

Simplify SWE reporting to single official row

Add intermediate SWE resolved-over-submitted metric

Use official SWE-bench resolved rate metric

Use markdown evaluation tables for reliable rendering

Clarify BFCL accuracy vs SWE submission coverage

Clarify quantization script as reference

Tighten model card accuracy (dates, deployment, metadata)

Update model card with standardized compression layout and BFCL/SWE results

Add BFCL v4 and SWE-bench Lite EveryEvalEver results

Append Every Eval Ever benchmark results table

Add every_eval_ever benchmark results

Update README.md

Create README.md

Add files using upload-large-folder tool

Add files using upload-large-folder tool

Add files using upload-large-folder tool

initial commit