feat: add benchmark command in Makefile

Author: S1M0N38

Makefile

@@ -1,5 +1,5 @@
 .DEFAULT_GOAL := help
-.PHONY: help install install-dev lint lint-fix format typecheck test test-cov quality clean setup teardown all
+.PHONY: help install install-dev lint lint-fix format typecheck test test-cov quality clean setup teardown all balatrobench
 
 # Colors for output
 YELLOW := \033[33m
@@ -91,3 +91,17 @@ teardown: ## Stop LiteLLM server and Balatro processes
 	@echo "$(YELLOW)Stopping Balatro...$(RESET)"
 	@./balatro.sh --kill 2>/dev/null || true
 	@echo "$(GREEN)✓ Services stopped$(RESET)"
+
+balatrobench: ## Run benchmark for all models and generate analysis
+	@echo "$(YELLOW)Starting benchmark runs for all models...$(RESET)"
+	@echo "$(YELLOW)Running cerebras/gpt-oss-120b...$(RESET)"
+	@balatrollm --runs-dir ./balatrobench --runs 3 --model cerebras/gpt-oss-120b || true
+	@echo "$(YELLOW)Running cerebras/qwen-3-235b-a22b-thinking-2507...$(RESET)"
+	@balatrollm --runs-dir ./balatrobench --runs 3 --model cerebras/qwen-3-235b-a22b-thinking-2507 || true
+	@echo "$(YELLOW)Running cerebras/qwen-3-235b-a22b-instruct-2507...$(RESET)"
+	@balatrollm --runs-dir ./balatrobench --runs 3 --model cerebras/qwen-3-235b-a22b-instruct-2507 || true
+	@echo "$(YELLOW)Running groq/openai/gpt-oss-20b...$(RESET)"
+	@balatrollm --runs-dir ./balatrobench --runs 3 --model groq/openai/gpt-oss-20b || true
+	@echo "$(YELLOW)Generating benchmark analysis...$(RESET)"
+	@balatrollm benchmark --runs-dir balatrobench/runs --output-dir balatrobench/benchmarks
+	@echo "$(GREEN)✓ Benchmark completed$(RESET)"