Pricing: Pay per second for GPUs and CPUs | Modal (https://modal.com/pricing) citeturn127view3 [wordlim: 200] Crawled: today; Content type: text/html; Source: open({"ref_id":"https://modal.com/pricing","lineno":null}); Total lines: 391 L0: 【0†Skip to content】 L1: L2: A new frontier for open-weight intelligence: Kimi K3.1 Pro is now available on Modal. A new frontier for open-weight intelligence: Kimi K3.1 Pro is now available on Modal. L3: L4: 【1†Learn more】 L5: L6: * Download logo as .svg L7: * 【2†Download wordmark as .svg】 L8: * * 【3†Go to press kit】 L9: L10: Product L11: 【4†Inference Serve LLMs, image, video, and audio models, from real-time to batch】 【5†Sandboxes Isolated environments for coding agents, RL rollouts, and untrusted code】 【6†Training Fine-tune, run RL, and evaluate models on GPUs that spin up in seconds】 L12: L13: 【7†Modal Infrastructure Infrastructure rebuilt for AI, from the container runtime and storage to a multi-cloud capacity pool】 L14: L15: Featured Model Endpoints L16: L17: 【8†Kimi K3.1 Pro】 【9†DeepSeek v4.1 Flash 1017】 【10†GLM 5.3】 【11†Muse Spark Open】 【12†See all models】 L18: Tools L19: L20: 【13†GPU Glossary A fast guide to GPU concepts】 【14†LLM Engine Advisor Pick the right model】 L21: L22: Resources L23: L24: 【15†Blog】 【16†Changelog】 【17†Startup Credits】 【18†Partners】 【19†Slack Community】 L25: L26: Company L27: L28: 【20†About】 【21†Careers】 【22†Events】 L29: L30: Featured L31: L32: 【23†Scaling to 1 million concurrent sandboxes in seconds Engineering】 L33: L34: * Product L35: L36: Product L37: 【4†Inference Serve LLMs, image, video, and audio models, from real-time to batch】 【5†Sandboxes Isolated environments for coding agents, RL rollouts, and untrusted code】 【6†Training Fine-tune, run RL, and evaluate models on GPUs that spin up in seconds】 【7†Modal Infrastructure Infrastructure rebuilt for AI, from the container runtime and storage to a multi-cloud capacity pool】 L38: L39: Featured Model Endpoints L40: L41: 【8†Kimi K3.1 Pro】 【9†DeepSeek v4.1 Flash 1017】 【10†GLM 5.3】 【11†Muse Spark Open】 【12†See all models】 L42: * Resources L43: L44: Tools L45: L46: 【13†GPU Glossary A fast guide to GPU concepts】 【14†LLM Engine Advisor Pick the right model】 L47: L48: Resources L49: L50: 【15†Blog】 【16†Changelog】 【17†Startup Credits】 【18†Partners】 【19†Slack Community】 L51: L52: Company L53: L54: 【20†About】 【21†Careers】 【22†Events】 L55: L56: Featured L57: L58: 【23†Scaling to 1 million concurrent sandboxes in seconds Engineering】 L59: L60: * 【24†Customers】 L61: * 【25†Pricing】 L62: * 【26†Docs】 L63: L64: 【27†Log in】 【28†Sign up】 L65: # Pricing as magical as our product L66: L67: With Modal, you always pay for what you use and nothing more. You never pay for idle resources — just actual compute time, by the CPU cycle. L68: L69: 【28†Get started】 【29†Contact us†portal.usepylon.com】 L70: ## Resource costs L71: L72: [Button: Per hour][Button: Per second] L73: L74: GPU Tasks L75: * Nvidia B300 $0.001972 0.0 0 1 9 7 2 / sec L76: * Nvidia B200 $0.001736 0.0 0 1 7 3 6 / sec L77: * Nvidia H200 SXM $0.001261 0.0 0 1 2 6 1 / sec L78: * Nvidia H100 SXM5 $0.001097 0.0 0 1 0 9 7 / sec L79: * Nvidia RTX PRO 6000 $0.000842 0.0 0 0 8 4 2 / sec L80: * Nvidia A100, 80 GB $0.000694 0.0 0 0 6 9 4 / sec L81: * Nvidia A100, 40 GB $0.000583 0.0 0 0 5 8 3 / sec L82: * Nvidia L40S $0.000542 0.0 0 0 5 4 2 / sec L83: * Nvidia A10 $0.000306 0.0 0 0 3 0 6 / sec L84: * Nvidia L4 $0.000222 0.0 0 0 2 2 2 / sec L85: * Nvidia T4 $0.000164 0.0 0 0 1 6 4 / sec L86: CPU $0.0000131 0.0 0 0 0 1 3 1 / core / sec L87: L88: Physical core (2 vCPU equivalent) L89: L90: *minimum of 0.125 cores per container L91: L92: Memory $0.00000222 0.0 0 0 0 0 2 2 2 / GiB / sec L93: L94: Volumes $0.09 0.0 9 / GiB / mo L95: L96: *includes 1 TiB / mo free L97: L98: Network Egress $0.04 0.0 4 / GiB L99: L100: 【30†*See plan for included usage / mo】 L101: ## Model APIs L102: L103: / 1M tokens L104: L105: Model L106: L107: Prompt Cached prompt Completion L108: * 【31†Kimi K3 Prompt:$3.00 Cached prompt:$0.30 Completion:$15.00 per 1M tokens. Try model】 L109: * 【32†Kimi K3.1 Pro Prompt:$2.00 Cached prompt:$0.20 Completion:$10.00 per 1M tokens. Try model】 L110: * 【33†Qwen3.8-Max Prompt:$2.00 Cached prompt:$0.25 Completion:$6.00 per 1M tokens. Try model】 L111: * 【34†Qwen3.8-Max VL Prompt:$2.00 Cached prompt:$0.25 Completion:$6.00 per 1M tokens. Try model】 L112: * 【35†Inkling NVFP4 Prompt:$1.20 Cached prompt:$0.27 Completion:$5.00 per 1M tokens. Try model】