Update README.md

README.md

@@ -46,6 +46,19 @@ For instance, serve the model as follows:
 vllm serve RedHatAI/Magistral-Small-2506-FP8 --tokenizer-mode mistral --config-format mistral --load-format mistral --tool-call-parser mistral --enable-auto-tool-choice
 ```
 
+## Evaluation
+
+GSM8k:
+
+```
+lm_eval --model local-completions --model_args model=RedHatAI/Magistral-Small-2506-FP8,base_url=http://0.0.0.0:9000/v1/completions,num_concurrent=500,tokenized_requests=False --tasks gsm8k --num_fewshot 5
+local-completions (model=RedHatAI/Magistral-Small-2506-FP8,base_url=http://0.0.0.0:9000/v1/completions,num_concurrent=500,tokenized_requests=False), gen_kwargs: (None), limit: None, num_fewshot: 5, batch_size: 1
+|Tasks|Version|     Filter     |n-shot|  Metric   |   |Value |   |Stderr|
+|-----|------:|----------------|-----:|-----------|---|-----:|---|-----:|
+|gsm8k|      3|flexible-extract|     5|exact_match|↑  |0.8923|±  |0.0085|
+|     |       |strict-match    |     5|exact_match|↑  |0.8886|±  |0.0087|
+```
+
 # Original Model Card
 
 Building upon Mistral Small 3.1 (2503), **with added reasoning capabilities**, undergoing SFT from Magistral Medium traces and RL on top, it's a small, efficient reasoning model with 24B parameters.