diff --git a/providers/novita.toml b/providers/novita.toml
new file mode 100644
index 0000000..2bb1a87
--- /dev/null
+++ b/providers/novita.toml
@@ -0,0 +1,91 @@
+# Novita AI — https://novita.ai
+# OpenAI-compatible LLM gateway. Endpoint: https://api.novita.ai/openai/v1
+# Pricing reflects values returned by GET /openai/v1/models on 2026-04-25
+# (input_token_price_per_m / output_token_price_per_m, scaled /10000 to USD).
+# Models: 6
+
+[provider]
+id = "novita"
+display_name = "Novita AI"
+api_key_env = "NOVITA_API_KEY"
+base_url = "https://api.novita.ai/openai/v1"
+key_required = true
+
+[[models]]
+id = "deepseek/deepseek-v3.2"
+display_name = "DeepSeek V3.2 (Novita)"
+tier = "frontier"
+context_window = 163840
+max_output_tokens = 65536
+input_cost_per_m = 0.269
+output_cost_per_m = 0.40
+supports_tools = true
+supports_vision = false
+supports_streaming = true
+aliases = []
+
+[[models]]
+id = "moonshotai/kimi-k2-thinking"
+display_name = "Kimi K2 Thinking (Novita)"
+tier = "frontier"
+context_window = 262144
+max_output_tokens = 262144
+input_cost_per_m = 0.60
+output_cost_per_m = 2.50
+supports_tools = true
+supports_vision = false
+supports_streaming = true
+supports_thinking = true
+aliases = []
+
+[[models]]
+id = "minimax/minimax-m2"
+display_name = "MiniMax M2 (Novita)"
+tier = "smart"
+context_window = 204800
+max_output_tokens = 131072
+input_cost_per_m = 0.30
+output_cost_per_m = 1.20
+supports_tools = true
+supports_vision = false
+supports_streaming = true
+aliases = []
+
+[[models]]
+id = "meta-llama/llama-3.3-70b-instruct"
+display_name = "Llama 3.3 70B (Novita)"
+tier = "smart"
+context_window = 131072
+max_output_tokens = 120000
+input_cost_per_m = 0.135
+output_cost_per_m = 0.40
+supports_tools = true
+supports_vision = false
+supports_streaming = true
+aliases = []
+
+[[models]]
+id = "qwen/qwen3-coder-30b-a3b-instruct"
+display_name = "Qwen3 Coder 30B (Novita)"
+tier = "balanced"
+context_window = 160000
+max_output_tokens = 32768
+input_cost_per_m = 0.07
+output_cost_per_m = 0.27
+supports_tools = true
+supports_vision = false
+supports_streaming = true
+aliases = []
+
+[[models]]
+id = "zai-org/glm-4.7-flash"
+display_name = "GLM 4.7 Flash (Novita)"
+tier = "balanced"
+context_window = 200000
+max_output_tokens = 128000
+input_cost_per_m = 0.07
+output_cost_per_m = 0.40
+supports_tools = true
+supports_vision = false
+supports_streaming = true
+aliases = []