{"data":[{"id":"phala/qwen-2.5-7b-instruct","name":"Qwen2.5 7B Instruct","created":1759532719,"input_modalities":["text"],"output_modalities":["text"],"context_length":32768,"max_output_length":32768,"pricing":{"prompt":"0.0000001","completion":"0.0000002"},"supported_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"supported_sampling_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","seed","stop","temperature","top_k","top_p"],"supported_features":["json_mode","structured_outputs","tools"],"hugging_face_id":"Qwen/Qwen2.5-7B-Instruct","is_tee":true,"providers":["phala"],"description":"Qwen2.5 7B is the latest series of Qwen large language models. Qwen2.5 brings the following improvements upon Qwen2:\n\n- Significantly more knowledge and has greatly improved capabilities in coding and mathematics, thanks to our specialized expert models in these domains.\n\n- Significant improvements in instruction following, generating long texts (over 8K tokens), understanding structured data (e.g, tables), and generating structured outputs especially JSON. More resilient to the diversity of system prompts, enhancing role-play implementation and condition-setting for chatbots.\n\n- Long-context Support up to 128K tokens and can generate up to 8K tokens.\n\n- Multilingual support for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.\n\nUsage of this model is subject to [Tongyi Qianwen LICENSE AGREEMENT](https://huggingface.co/Qwen/Qwen1.5-110B-Chat/blob/main/LICENSE).","metadata":{}},{"id":"phala/kev-4b","name":"Jared Palmer: Kev 4B","created":1791855524,"input_modalities":["text"],"output_modalities":["decisions"],"context_length":8192,"max_output_length":8192,"pricing":{"prompt":"0.000000042","completion":"0"},"supported_parameters":[],"supported_sampling_parameters":[],"supported_features":[],"hugging_face_id":"jaredpalmer/kev-4b","is_tee":true,"providers":["phala"],"description":"Kev 4B is an open-weight System One model from Jared Palmer, fine-tuned from Qwen3.5 4B Base. It makes fast, structured decisions for software: given a state and typed questions, it returns calibrated answers instead of text, as a yes/no probability (noul), a pick among options (choice), or a level on a scale (score). Call it through POST /v1/systemone, compatible with the TypeSafe SDK; it does not serve chat. Served on Phala in a TDX-attested enclave with an 8,192-token context.","metadata":{}},{"id":"phala/qwen3.8-27b","name":"Qwen: Qwen3.8 27B","created":1787560779,"input_modalities":["text","image","video"],"output_modalities":["text"],"context_length":1000000,"max_output_length":262144,"pricing":{"prompt":"0.0000002","completion":"0.0000025","input_cache_read":"0.00000005"},"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","presence_penalty","reasoning","reasoning_effort","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"supported_sampling_parameters":["frequency_penalty","logit_bias","max_tokens","presence_penalty","repetition_penalty","seed","stop","temperature","top_k","top_p"],"supported_features":["json_mode","logprobs","reasoning","structured_outputs","tools"],"reasoning":{"mandatory":false,"default_enabled":true},"hugging_face_id":"Qwen/Qwen3.8-27B","is_tee":true,"providers":["chutes","phala"],"description":"Qwen3.8 27B is an open-weight dense vision-language model from Qwen. It is suited for coding, professional workflows, research, multimodal interaction, and long-running agent tasks, with flexible thinking that can be enabled or disabled. Served on Phala in a TDX-attested enclave.","metadata":{}},{"id":"phala/kimi-k3.1-pro","name":"MoonshotAI: Kimi K3.1 Pro","created":1793810541,"input_modalities":["text","image","video"],"output_modalities":["text"],"context_length":1048576,"max_output_length":1048576,"pricing":{"prompt":"0.000002","completion":"0.00001","input_cache_read":"0.0000002"},"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","reasoning_effort","repetition_penalty","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"supported_sampling_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","stop","temperature","top_k","top_p"],"supported_features":["json_mode","logprobs","reasoning","structured_outputs","tools"],"reasoning":{"mandatory":false,"default_enabled":true},"hugging_face_id":"moonshotai/Kimi-K3.1-Pro","is_tee":true,"providers":["phala","tinfoil"],"description":"Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. It is built for long-horizon coding, 3D creation, financial research, and end-to-end agentic knowledge work, with native image and video understanding.","metadata":{}},{"id":"phala/gemma-4-26b-a4b-uncensored","name":"Phala: Gemma-4 26B-A4B Uncensored (Heretic)","created":1779562961,"input_modalities":["text","image"],"output_modalities":["text"],"context_length":262144,"max_output_length":262144,"pricing":{"prompt":"0.00000015","completion":"0.0000007","input_cache_read":"0.00000005"},"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"supported_sampling_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","seed","stop","temperature","top_k","top_p"],"supported_features":["json_mode","logprobs","reasoning","structured_outputs","tools"],"reasoning":{"mandatory":false,"default_enabled":false},"is_tee":true,"providers":["phala"],"description":"Uncensored \"Heretic\" variant of google/gemma-4-26B-A4B-it created using Heretic v1.2.0 with the Arbitrary-Rank Ablation (ARA) method and row-norm preservation. Refusals drop from 100/100 to 11/100 with KL divergence 0.0499 vs the base model. The base Gemma 4 26B A4B is a Mixture-of-Experts model with 25.2B total / 3.8B active parameters (8 active / 128 total experts), 30-layer transformer with hybrid local sliding (1024) + global attention, supporting a 256K context window. Natively multimodal (text + images, variable aspect ratios). Strong on coding, reasoning, function calling, with native system prompt support across 35+ languages. Served on Phala in TDX-attested H200 enclave with end-to-end ECDSA response signing; vLLM-compatible FP8-Static quantization by cloud19 (router excluded from quantization).","metadata":{}},{"id":"phala/qwen3.8-27b-uncensored","name":"Phala: Qwen3.8 27B Uncensored","created":1789549583,"input_modalities":["text","image","video"],"output_modalities":["text"],"context_length":1000000,"max_output_length":262144,"pricing":{"prompt":"0.0000003","completion":"0.0000015","input_cache_read":"0.00000005"},"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","reasoning_effort","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"supported_sampling_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","seed","stop","temperature","top_k","top_p"],"supported_features":["json_mode","logprobs","reasoning","structured_outputs","tools"],"reasoning":{"mandatory":false,"default_enabled":true},"hugging_face_id":"PhalaCloud/Qwen3.8-27B-Uncensored-LoRA","is_tee":true,"providers":["phala"],"description":"Uncensored variant of Qwen3.8 27B, built as a lightweight abliteration adapter on the official FP8 weights. Refuses far less while keeping the base model's capabilities, tool calling, and image and video input. Served with SGLang in a TDX-attested enclave, with a 1M-token context.","metadata":{}},{"id":"phala/deepseek-v4.1-flash-1017-uncensored","name":"Phala: DeepSeek V4.1 Flash 1017 Uncensored","created":1794968202,"input_modalities":["text","image"],"output_modalities":["text"],"context_length":1048576,"max_output_length":262144,"pricing":{"prompt":"0.0000003","completion":"0.0000012","input_cache_read":"0.000000006"},"supported_parameters":["frequency_penalty","include_reasoning","logprobs","max_tokens","presence_penalty","reasoning","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"supported_sampling_parameters":["frequency_penalty","max_tokens","presence_penalty","stop","temperature","top_k","top_p"],"supported_features":["json_mode","logprobs","reasoning","structured_outputs","tools"],"reasoning":{"mandatory":false,"default_enabled":true},"hugging_face_id":"PhalaCloud/DeepSeek-V4.1-Flash-1017-Uncensored-LoRA","is_tee":true,"providers":["phala"],"description":"Uncensored variant of DeepSeek V4.1 Flash 1017, built as a lightweight LoRA adapter served unmerged on the official FP8 weights. Refuses far less while keeping the base model's capabilities, tool calling, image input and 1M-token context. Served with SGLang in a TDX-attested enclave with NVIDIA confidential computing. Reasoning is on by default and can be turned off or capped per request.","metadata":{}}]}