{"data":[{"id":"moonshot-kimi-k3","name":"kimi-k3","display_name":"Kimi K3","description":"A flagship reasoning-capable multimodal LLM with 2.8 trillion parameters, built on Kimi Delta Attention (a hybrid linear attention mechanism) with native visual understanding and tool-use support.","creator":"moonshot","family":"kimi","tier":"","version":"k3","type":"language","size_in_bn":2779.932,"modalities":{"input":["image","pdf","text","video"],"output":["text"]},"context_window":1048576,"max_output_tokens":943718,"tool_use_system_prompt_tokens":0,"output_vector_sizes":[],"knowledge_cutoff":null,"training_data_cutoff":null,"supported_reasoning_efforts":["default"],"tokenizer":"Other","capabilities":{"function_calling":true,"parallel_function_calling":true,"structured_outputs":true,"prompt_caching":true,"reasoning":true,"web_search":true,"computer_use":false,"code_execution":false,"file_search":false,"url_context":false,"assistant_prefill":false,"native_structured_output":true,"adaptive_reasoning":false},"release_date":"2026-07-16","earliest_deprecation_date":null,"deprecated":false,"has_pricing":true,"provider_count":15,"ids":["accounts/fireworks/models/kimi-k3","aihubmix/kimi-k3","azure_ai/FW-Kimi-K3","databricks/databricks-kimi-k3","deepinfra/moonshotai/Kimi-K3","fireworks_ai/accounts/fireworks/models/kimi-k3","fireworks_ai/kimi-k3","global.moonshotai.kimi-k3","kimi-k3","kimi-k3-low","moonshot-kimi-k3","moonshot/kimi-k3","moonshotai.kimi-k3","moonshotai/kimi-k3","moonshotai/Kimi-K3","nebius/moonshotai/Kimi-K3","novita/moonshotai/kimi-k3","openrouter/moonshotai/kimi-k3","perplexity/perplexity/kimi-k3","together_ai/moonshotai/Kimi-K3","us.moonshotai.kimi-k3"],"hf_likes":7099,"hf_downloads":2850,"hf_downloads_all_time":2855,"hf_trending_score":6642,"updated_at":"2026-09-22 08:02:55"},{"id":"moonshot-kimi-k2-5-non-reasoning","name":"kimi-k2-5-non-reasoning","display_name":"Kimi K2.5 Non-Reasoning","description":"A non-chain-of-thought variant of Kimi K2.5, optimized for fast standard inference without extended reasoning steps.","creator":"moonshot","family":"kimi","tier":"","version":"k2-5","type":"","size_in_bn":null,"modalities":{"input":[],"output":[]},"context_window":null,"max_output_tokens":null,"tool_use_system_prompt_tokens":0,"output_vector_sizes":[],"knowledge_cutoff":null,"training_data_cutoff":null,"supported_reasoning_efforts":[],"tokenizer":null,"capabilities":{"function_calling":false,"parallel_function_calling":false,"structured_outputs":false,"prompt_caching":false,"reasoning":false,"web_search":false,"computer_use":false,"code_execution":false,"file_search":false,"url_context":false,"assistant_prefill":false,"native_structured_output":false,"adaptive_reasoning":false},"release_date":null,"earliest_deprecation_date":null,"deprecated":false,"has_pricing":false,"provider_count":0,"ids":["kimi-k2-5-non-reasoning","moonshot-kimi-k2-5-non-reasoning"],"hf_likes":null,"hf_downloads":null,"hf_downloads_all_time":null,"hf_trending_score":null,"updated_at":"2026-04-20 16:51:01"},{"id":"moonshot-kimi-linear-48b-a3b-instruct","name":"kimi-linear-48b-a3b-instruct","display_name":"Kimi Linear 48B A3B Instruct","description":"A large mixture-of-experts LLM from Kimi with linear attention architecture, optimized for long-context instruction following at 48B total parameters with 3B active.","creator":"moonshot","family":"kimi","tier":"","version":null,"type":"language","size_in_bn":48,"modalities":{"input":["text"],"output":["text"]},"context_window":null,"max_output_tokens":null,"tool_use_system_prompt_tokens":0,"output_vector_sizes":[],"knowledge_cutoff":null,"training_data_cutoff":null,"supported_reasoning_efforts":[],"tokenizer":null,"capabilities":{"function_calling":false,"parallel_function_calling":false,"structured_outputs":false,"prompt_caching":false,"reasoning":false,"web_search":false,"computer_use":false,"code_execution":false,"file_search":false,"url_context":false,"assistant_prefill":false,"native_structured_output":false,"adaptive_reasoning":false},"release_date":null,"earliest_deprecation_date":null,"deprecated":false,"has_pricing":false,"provider_count":0,"ids":["kimi-linear-48b-a3b-instruct","moonshot-kimi-linear-48b-a3b-instruct"],"hf_likes":null,"hf_downloads":null,"hf_downloads_all_time":null,"hf_trending_score":null,"updated_at":"2026-09-22 08:02:55"},{"id":"moonshot-kimi-k2-instruct-0905-v0","name":"kimi-k2-instruct-0905-v0","display_name":"Kimi K2 Instruct V0","description":"A speculative decoding draft model (EAGLE variant) for the Kimi K2 Instruct model, enabling faster inference through token prediction.","creator":"moonshot","family":"kimi","tier":"","version":"k2","type":"language","size_in_bn":null,"modalities":{"input":["text"],"output":["text"]},"context_window":null,"max_output_tokens":null,"tool_use_system_prompt_tokens":0,"output_vector_sizes":[],"knowledge_cutoff":null,"training_data_cutoff":null,"supported_reasoning_efforts":[],"tokenizer":null,"capabilities":{"function_calling":false,"parallel_function_calling":false,"structured_outputs":false,"prompt_caching":false,"reasoning":false,"web_search":false,"computer_use":false,"code_execution":false,"file_search":false,"url_context":false,"assistant_prefill":false,"native_structured_output":false,"adaptive_reasoning":false},"release_date":null,"earliest_deprecation_date":null,"deprecated":false,"has_pricing":false,"provider_count":0,"ids":["accounts/fireworks/models/eagle1-kimi-k2-instruct-0905-v0","accounts/fireworks/models/eagle3-kimi-k2-instruct-0905-v0","moonshot-kimi-k2-instruct-0905-v0"],"hf_likes":null,"hf_downloads":null,"hf_downloads_all_time":null,"hf_trending_score":null,"updated_at":"2026-09-22 08:02:55"},{"id":"moonshot-kimi-k2-turbo","name":"kimi-k2-turbo","display_name":"Kimi K2 Turbo","description":"A high-throughput variant of Kimi K2 delivering up to 100 tokens per second, optimized for speed-critical tool-use and conversational applications.","creator":"moonshot","family":"kimi","tier":"","version":"k2","type":"language","size_in_bn":null,"modalities":{"input":["text"],"output":["text"]},"context_window":256000,"max_output_tokens":16384,"tool_use_system_prompt_tokens":0,"output_vector_sizes":[],"knowledge_cutoff":null,"training_data_cutoff":null,"supported_reasoning_efforts":[],"tokenizer":null,"capabilities":{"function_calling":true,"parallel_function_calling":false,"structured_outputs":false,"prompt_caching":false,"reasoning":false,"web_search":false,"computer_use":false,"code_execution":false,"file_search":false,"url_context":false,"assistant_prefill":false,"native_structured_output":false,"adaptive_reasoning":false},"release_date":"2025-09-05","earliest_deprecation_date":null,"deprecated":false,"has_pricing":false,"provider_count":0,"ids":["moonshot-kimi-k2-turbo","moonshotai/kimi-k2-turbo"],"hf_likes":null,"hf_downloads":null,"hf_downloads_all_time":null,"hf_trending_score":null,"updated_at":"2026-09-22 08:02:55"},{"id":"moonshot-kimi-k2-5-fp4","name":"kimi-k2-5-fp4","display_name":"Kimi K2.5 FP4","description":"A large language model in the Kimi K2 series quantized to FP4 precision for reduced memory footprint and faster inference.","creator":"moonshot","family":"kimi","tier":"","version":"k2-5","type":"language","size_in_bn":null,"modalities":{"input":["text"],"output":["text"]},"context_window":262144,"max_output_tokens":null,"tool_use_system_prompt_tokens":0,"output_vector_sizes":[],"knowledge_cutoff":null,"training_data_cutoff":null,"supported_reasoning_efforts":[],"tokenizer":null,"capabilities":{"function_calling":false,"parallel_function_calling":false,"structured_outputs":false,"prompt_caching":false,"reasoning":false,"web_search":false,"computer_use":false,"code_execution":false,"file_search":false,"url_context":false,"assistant_prefill":false,"native_structured_output":false,"adaptive_reasoning":false},"release_date":null,"earliest_deprecation_date":null,"deprecated":false,"has_pricing":true,"provider_count":1,"ids":["moonshot-kimi-k2-5-fp4","together_ai/moonshotai/Kimi-K2.5-fp4"],"hf_likes":null,"hf_downloads":null,"hf_downloads_all_time":null,"hf_trending_score":null,"updated_at":"2026-09-22 08:02:55"},{"id":"moonshot-kimi-k2-5-thinking","name":"moonshot-kimi-k2-5-thinking","display_name":"Kimi K2.5 Thinking","description":"A thinking-mode variant of Kimi K2.5 that performs step-by-step chain-of-thought reasoning for complex problem solving.","creator":"moonshot","family":"kimi","tier":"","version":"k2-5","type":"language","size_in_bn":null,"modalities":{"input":["text"],"output":["text"]},"context_window":null,"max_output_tokens":null,"tool_use_system_prompt_tokens":0,"output_vector_sizes":[],"knowledge_cutoff":null,"training_data_cutoff":null,"supported_reasoning_efforts":[],"tokenizer":null,"capabilities":{"function_calling":false,"parallel_function_calling":false,"structured_outputs":false,"prompt_caching":false,"reasoning":false,"web_search":false,"computer_use":false,"code_execution":false,"file_search":false,"url_context":false,"assistant_prefill":false,"native_structured_output":false,"adaptive_reasoning":false},"release_date":null,"earliest_deprecation_date":null,"deprecated":false,"has_pricing":true,"provider_count":1,"ids":["moonshot-kimi-k2-5-thinking"],"hf_likes":null,"hf_downloads":null,"hf_downloads_all_time":null,"hf_trending_score":null,"updated_at":"2026-09-22 08:02:55"},{"id":"moonshot-kimi-k2-6-fast","name":"kimi-k2-6-fast","display_name":"Kimi K2.6 Fast","description":"A fast-tier variant of Moonshot's Kimi K2.6 LLM, optimized for low-latency responses in agentic and tool-use scenarios.","creator":"moonshot","family":"kimi","tier":"","version":"k2-6","type":"language","size_in_bn":null,"modalities":{"input":["image"],"output":["text"]},"context_window":262144,"max_output_tokens":32768,"tool_use_system_prompt_tokens":0,"output_vector_sizes":[],"knowledge_cutoff":null,"training_data_cutoff":null,"supported_reasoning_efforts":["default"],"tokenizer":null,"capabilities":{"function_calling":true,"parallel_function_calling":false,"structured_outputs":true,"prompt_caching":false,"reasoning":true,"web_search":false,"computer_use":false,"code_execution":false,"file_search":false,"url_context":false,"assistant_prefill":false,"native_structured_output":false,"adaptive_reasoning":false},"release_date":null,"earliest_deprecation_date":null,"deprecated":false,"has_pricing":true,"provider_count":1,"ids":["fireworks_ai/accounts/fireworks/routers/kimi-k2p6-fast","fireworks_ai/kimi-k2p6-fast","moonshot-kimi-k2-6-fast"],"hf_likes":null,"hf_downloads":null,"hf_downloads_all_time":null,"hf_trending_score":null,"updated_at":"2026-09-22 08:02:55"},{"id":"moonshot-kimi-k2-6-thinking","name":"moonshot-kimi-k2-6-thinking","display_name":"Kimi K2.6 Thinking","description":"A reasoning-focused LLM in the Kimi K2 series with extended thinking capabilities for complex multi-step problem solving.","creator":"moonshot","family":"kimi","tier":"","version":"k2-6","type":"language","size_in_bn":null,"modalities":{"input":["text"],"output":["text"]},"context_window":null,"max_output_tokens":null,"tool_use_system_prompt_tokens":0,"output_vector_sizes":[],"knowledge_cutoff":null,"training_data_cutoff":null,"supported_reasoning_efforts":[],"tokenizer":null,"capabilities":{"function_calling":false,"parallel_function_calling":false,"structured_outputs":false,"prompt_caching":false,"reasoning":false,"web_search":false,"computer_use":false,"code_execution":false,"file_search":false,"url_context":false,"assistant_prefill":false,"native_structured_output":false,"adaptive_reasoning":false},"release_date":null,"earliest_deprecation_date":null,"deprecated":false,"has_pricing":true,"provider_count":1,"ids":["moonshot-kimi-k2-6-thinking"],"hf_likes":null,"hf_downloads":null,"hf_downloads_all_time":null,"hf_trending_score":null,"updated_at":"2026-09-22 08:02:55"},{"id":"moonshot-kimi-k2-7-code-fast","name":"kimi-k2-7-code-fast","display_name":"Kimi K2.7 Code Fast","description":"A fast-tier code-specialized variant of Moonshot's Kimi K2.7 LLM, designed for rapid code generation and programming assistance.","creator":"moonshot","family":"kimi","tier":"","version":"k2-7","type":"language","size_in_bn":null,"modalities":{"input":["image"],"output":["text"]},"context_window":262144,"max_output_tokens":32768,"tool_use_system_prompt_tokens":0,"output_vector_sizes":[],"knowledge_cutoff":null,"training_data_cutoff":null,"supported_reasoning_efforts":["default"],"tokenizer":null,"capabilities":{"function_calling":true,"parallel_function_calling":false,"structured_outputs":true,"prompt_caching":false,"reasoning":true,"web_search":false,"computer_use":false,"code_execution":false,"file_search":false,"url_context":false,"assistant_prefill":false,"native_structured_output":false,"adaptive_reasoning":false},"release_date":null,"earliest_deprecation_date":null,"deprecated":false,"has_pricing":true,"provider_count":1,"ids":["fireworks_ai/accounts/fireworks/routers/kimi-k2p7-code-fast","fireworks_ai/kimi-k2p7-code-fast","moonshot-kimi-k2-7-code-fast"],"hf_likes":null,"hf_downloads":null,"hf_downloads_all_time":null,"hf_trending_score":null,"updated_at":"2026-09-22 08:02:55"},{"id":"moonshot-kimi-k2-7-code-highspeed","name":"kimi-k2-7-code-highspeed","display_name":"Kimi K2.7 Code HighSpeed","description":"A code-specialized variant of Kimi K2.7 optimized for high-throughput generation, delivering fast token output for latency-sensitive coding, reasoning, tool-use, and vision tasks.","creator":"moonshot","family":"kimi","tier":"","version":"k2-7","type":"language","size_in_bn":null,"modalities":{"input":["image","pdf"],"output":["text"]},"context_window":262144,"max_output_tokens":32768,"tool_use_system_prompt_tokens":0,"output_vector_sizes":[],"knowledge_cutoff":null,"training_data_cutoff":null,"supported_reasoning_efforts":["default"],"tokenizer":null,"capabilities":{"function_calling":true,"parallel_function_calling":false,"structured_outputs":true,"prompt_caching":true,"reasoning":true,"web_search":false,"computer_use":false,"code_execution":false,"file_search":false,"url_context":false,"assistant_prefill":false,"native_structured_output":false,"adaptive_reasoning":false},"release_date":"2026-06-15","earliest_deprecation_date":null,"deprecated":false,"has_pricing":true,"provider_count":2,"ids":["aihubmix/kimi-k2.7-code-highspeed","moonshot-kimi-k2-7-code-highspeed","moonshotai/kimi-k2.7-code-highspeed"],"hf_likes":null,"hf_downloads":null,"hf_downloads_all_time":null,"hf_trending_score":null,"updated_at":"2026-09-22 08:02:55"},{"id":"moonshot-kimi-k3-batch","name":"kimi-k3-batch","display_name":"Kimi K3 Batch","description":"A batch-processing variant of Moonshot AI's Kimi K3 large language model optimized for asynchronous high-throughput workloads.","creator":"moonshot","family":"kimi","tier":"","version":"k3","type":"language","size_in_bn":null,"modalities":{"input":["image"],"output":["text"]},"context_window":1048576,"max_output_tokens":943718,"tool_use_system_prompt_tokens":0,"output_vector_sizes":[],"knowledge_cutoff":null,"training_data_cutoff":null,"supported_reasoning_efforts":["default"],"tokenizer":null,"capabilities":{"function_calling":true,"parallel_function_calling":false,"structured_outputs":true,"prompt_caching":true,"reasoning":true,"web_search":false,"computer_use":false,"code_execution":false,"file_search":false,"url_context":false,"assistant_prefill":false,"native_structured_output":false,"adaptive_reasoning":false},"release_date":null,"earliest_deprecation_date":null,"deprecated":false,"has_pricing":true,"provider_count":1,"ids":["moonshot-kimi-k3-batch","openrouter/moonshotai/kimi-k3:batch"],"hf_likes":null,"hf_downloads":null,"hf_downloads_all_time":null,"hf_trending_score":null,"updated_at":"2026-09-22 08:02:55"},{"id":"moonshot-kimi-k3-coding","name":"kimi-k3-coding","display_name":"Kimi K3 Coding","description":"A code-specialized variant of the Kimi K3 LLM tuned for software development tasks such as code generation and debugging.","creator":"moonshot","family":"kimi","tier":"","version":"k3","type":"language","size_in_bn":null,"modalities":{"input":["image"],"output":["text"]},"context_window":1048576,"max_output_tokens":null,"tool_use_system_prompt_tokens":0,"output_vector_sizes":[],"knowledge_cutoff":null,"training_data_cutoff":null,"supported_reasoning_efforts":["default"],"tokenizer":null,"capabilities":{"function_calling":true,"parallel_function_calling":false,"structured_outputs":true,"prompt_caching":true,"reasoning":true,"web_search":false,"computer_use":false,"code_execution":false,"file_search":false,"url_context":false,"assistant_prefill":false,"native_structured_output":false,"adaptive_reasoning":false},"release_date":null,"earliest_deprecation_date":null,"deprecated":false,"has_pricing":true,"provider_count":1,"ids":["aihubmix/coding-kimi-k3","moonshot-kimi-k3-coding"],"hf_likes":null,"hf_downloads":null,"hf_downloads_all_time":null,"hf_trending_score":null,"updated_at":"2026-09-22 08:02:55"},{"id":"moonshot-kimi-k3-fast","name":"kimi-k3-fast","display_name":"Kimi K3 Fast","description":"A speed-optimized variant of Moonshot AI's Kimi K3 large language model, tuned for lower-latency responses.","creator":"moonshot","family":"kimi","tier":"","version":"k3","type":"language","size_in_bn":null,"modalities":{"input":["image"],"output":["text"]},"context_window":1048576,"max_output_tokens":131072,"tool_use_system_prompt_tokens":0,"output_vector_sizes":[],"knowledge_cutoff":null,"training_data_cutoff":null,"supported_reasoning_efforts":["default"],"tokenizer":null,"capabilities":{"function_calling":true,"parallel_function_calling":false,"structured_outputs":true,"prompt_caching":false,"reasoning":true,"web_search":false,"computer_use":false,"code_execution":false,"file_search":false,"url_context":false,"assistant_prefill":false,"native_structured_output":false,"adaptive_reasoning":false},"release_date":null,"earliest_deprecation_date":null,"deprecated":false,"has_pricing":true,"provider_count":1,"ids":["fireworks_ai/accounts/fireworks/routers/kimi-k3-fast","fireworks_ai/kimi-k3-fast","moonshot-kimi-k3-fast"],"hf_likes":null,"hf_downloads":null,"hf_downloads_all_time":null,"hf_trending_score":null,"updated_at":"2026-09-22 08:02:55"},{"id":"moonshot-kimi-k3-us","name":"kimi-k3-us","display_name":"Kimi K3 US","description":"A US-region deployment of Moonshot AI's Kimi K3 large language model for regional serving.","creator":"moonshot","family":"kimi","tier":"","version":"k3","type":"language","size_in_bn":null,"modalities":{"input":["image"],"output":["text"]},"context_window":1048576,"max_output_tokens":131072,"tool_use_system_prompt_tokens":0,"output_vector_sizes":[],"knowledge_cutoff":null,"training_data_cutoff":null,"supported_reasoning_efforts":["default"],"tokenizer":null,"capabilities":{"function_calling":true,"parallel_function_calling":false,"structured_outputs":true,"prompt_caching":false,"reasoning":true,"web_search":false,"computer_use":false,"code_execution":false,"file_search":false,"url_context":false,"assistant_prefill":false,"native_structured_output":false,"adaptive_reasoning":false},"release_date":null,"earliest_deprecation_date":null,"deprecated":false,"has_pricing":true,"provider_count":1,"ids":["fireworks_ai/accounts/fireworks/routers/kimi-k3-us","fireworks_ai/kimi-k3-us","moonshot-kimi-k3-us"],"hf_likes":null,"hf_downloads":null,"hf_downloads_all_time":null,"hf_trending_score":null,"updated_at":"2026-09-22 08:02:55"}],"pagination":{"page_size":50,"has_next":false,"next_token":null,"total_count":15},"meta":{"updated_at":"2026-09-22","request_id":"8866c65a-bc5a-4491-9623-c54afce9752c","execution_ms":8}}