feat(novita-ai): refresh and audit model catalog

models/alibaba/qwen-mt-plus.toml

@@ -0,0 +1,22 @@
+# https://www.qwencloud.com/models/qwen-mt-plus
+# https://www.alibabacloud.com/help/en/model-studio/qwen-mt-plus
+# https://qwenlm.github.io/blog/qwen-mt/
+# https://novita.ai/models/model-detail/qwen-qwen-mt-plus
+name = "Qwen MT Plus"
+description = "Flagship Qwen translation model with terminology control and support for 92 languages"
+family = "qwen"
+release_date = "2025-07-24"
+last_updated = "2025-07-24"
+attachment = false
+reasoning = false
+tool_call = false
+temperature = true
+open_weights = false
+
+[limit]
+context = 16_384
+output = 8_192
+
+[modalities]
+input = ["text"]
+output = ["text"]

models/alibaba/qwen3-embedding-8b.toml

@@ -0,0 +1,26 @@
+# https://huggingface.co/Qwen/Qwen3-Embedding-8B
+# https://qwenlm.github.io/blog/qwen3-embedding/
+name = "Qwen3 Embedding 8B"
+description = "Multilingual text embedding model for retrieval, code search, classification, and clustering"
+type = "embedding"
+family = "qwen"
+release_date = "2025-06-05"
+last_updated = "2025-06-05"
+attachment = false
+reasoning = false
+tool_call = false
+temperature = false
+open_weights = true
+license = "Apache-2.0"
+
+[limit]
+context = 32_768
+output = 0
+
+[modalities]
+input = ["text"]
+output = ["text"]
+
+[[weights]]
+label = "Hugging Face"
+url = "https://huggingface.co/Qwen/Qwen3-Embedding-8B"

models/alibaba/qwen3-omni-30b-a3b-thinking.toml

@@ -0,0 +1,25 @@
+# https://huggingface.co/Qwen/Qwen3-Omni-30B-A3B-Thinking
+# https://qwen.ai/blog?id=qwen3-omni
+name = "Qwen3 Omni 30B A3B Thinking"
+description = "Multimodal reasoning model accepting text, images, video, and audio with text output"
+family = "qwen"
+release_date = "2025-09-22"
+last_updated = "2025-09-22"
+attachment = true
+reasoning = true
+tool_call = true
+temperature = true
+open_weights = true
+license = "Apache-2.0"
+
+[limit]
+context = 65_536
+output = 16_384
+
+[modalities]
+input = ["text", "image", "video", "audio"]
+output = ["text"]
+
+[[weights]]
+label = "Hugging Face"
+url = "https://huggingface.co/Qwen/Qwen3-Omni-30B-A3B-Thinking"

models/alibaba/qwen3-reranker-8b.toml

@@ -0,0 +1,26 @@
+# https://huggingface.co/Qwen/Qwen3-Reranker-8B
+# https://qwenlm.github.io/blog/qwen3-embedding/
+name = "Qwen3 Reranker 8B"
+description = "Multilingual query-document reranking model supporting more than 100 languages"
+type = "reranking"
+family = "qwen"
+release_date = "2025-06-05"
+last_updated = "2025-06-05"
+attachment = false
+reasoning = false
+tool_call = false
+temperature = false
+open_weights = true
+license = "Apache-2.0"
+
+[limit]
+context = 32_768
+output = 0
+
+[modalities]
+input = ["text"]
+output = ["text"]
+
+[[weights]]
+label = "Hugging Face"
+url = "https://huggingface.co/Qwen/Qwen3-Reranker-8B"

models/alibaba/qwen3-vl-30b-a3b-instruct.toml

@@ -0,0 +1,25 @@
+# https://huggingface.co/Qwen/Qwen3-VL-30B-A3B-Instruct
+# https://github.com/QwenLM/Qwen3-VL (News: 2025.10.4)
+name = "Qwen3 VL 30B A3B Instruct"
+description = "Open vision-language MoE model for image and video understanding, documents, and visual agents"
+family = "qwen"
+release_date = "2025-10-04"
+last_updated = "2025-10-04"
+attachment = true
+reasoning = false
+tool_call = true
+temperature = true
+open_weights = true
+license = "Apache-2.0"
+
+[limit]
+context = 262_144
+output = 32_768
+
+[modalities]
+input = ["text", "image", "video"]
+output = ["text"]
+
+[[weights]]
+label = "Hugging Face"
+url = "https://huggingface.co/Qwen/Qwen3-VL-30B-A3B-Instruct"

models/apodex/apodex-1.1-mini.toml

@@ -0,0 +1,25 @@
+# https://www.apodex.com/blog/apodex-1.1-scaling-agentic-intelligence-for-complex-work
+# https://huggingface.co/apodex/Apodex-1.1-mini
+# Output limit: https://novita.ai/models/model-detail/apodex-apodex-1.1-mini
+name = "Apodex 1.1 Mini"
+description = "Open-weight 35B MoE model for long-horizon research, forecasting, and tool-assisted work"
+release_date = "2026-08-24"
+last_updated = "2026-08-24"
+attachment = false
+reasoning = true
+tool_call = true
+temperature = true
+open_weights = true
+license = "Apache-2.0"
+
+[limit]
+context = 262_144
+output = 262_144
+
+[modalities]
+input = ["text"]
+output = ["text"]
+
+[[weights]]
+label = "Hugging Face"
+url = "https://huggingface.co/apodex/Apodex-1.1-mini"

models/baai/bge-m3.toml

@@ -0,0 +1,26 @@
+# https://huggingface.co/BAAI/bge-m3
+# https://github.com/FlagOpen/FlagEmbedding/tree/master/research/BGE_M3
+name = "BGE M3"
+description = "Multilingual embedding model combining dense, sparse, and multi-vector retrieval"
+type = "embedding"
+family = "bge"
+release_date = "2024-01-30"
+last_updated = "2024-01-30"
+attachment = false
+reasoning = false
+tool_call = false
+temperature = false
+open_weights = true
+license = "MIT"
+
+[limit]
+context = 8_192
+output = 0
+
+[modalities]
+input = ["text"]
+output = ["text"]
+
+[[weights]]
+label = "Hugging Face"
+url = "https://huggingface.co/BAAI/bge-m3"

models/baai/bge-reranker-v2-m3.toml

@@ -0,0 +1,25 @@
+# https://huggingface.co/BAAI/bge-reranker-v2-m3
+name = "BGE Reranker V2 M3"
+description = "Lightweight multilingual cross-encoder that scores query-document relevance"
+type = "reranking"
+family = "bge"
+release_date = "2024-03-15"
+last_updated = "2024-03-15"
+attachment = false
+reasoning = false
+tool_call = false
+temperature = false
+open_weights = true
+license = "Apache-2.0"
+
+[limit]
+context = 8_192
+output = 0
+
+[modalities]
+input = ["text"]
+output = ["text"]
+
+[[weights]]
+label = "Hugging Face"
+url = "https://huggingface.co/BAAI/bge-reranker-v2-m3"

models/baichuan/baichuan-m2-32b.toml

@@ -0,0 +1,24 @@
+# https://huggingface.co/baichuan-inc/Baichuan-M2-32B
+name = "Baichuan M2 32B"
+description = "Open medical model trained for clinical reasoning and patient interaction with on, off, and auto thinking modes"
+family = "baichuan"
+release_date = "2025-08-13"
+last_updated = "2025-08-13"
+attachment = false
+reasoning = true
+tool_call = false
+temperature = true
+open_weights = true
+license = "Apache-2.0"
+
+[limit]
+context = 131_072
+output = 131_072
+
+[modalities]
+input = ["text"]
+output = ["text"]
+
+[[weights]]
+label = "Hugging Face"
+url = "https://huggingface.co/baichuan-inc/Baichuan-M2-32B"

models/baidu/cobuddy.toml

@@ -0,0 +1,19 @@
+# https://openrouter.ai/baidu/cobuddy
+# https://novita.ai/models/model-detail/baidu-cobuddy
+name = "CoBuddy"
+description = "Baidu Qianfan model optimized for code generation and tool-using coding agents"
+release_date = "2026-05-06"
+last_updated = "2026-05-06"
+attachment = false
+reasoning = true
+tool_call = true
+temperature = true
+open_weights = false
+
+[limit]
+context = 131_072
+output = 65_536
+
+[modalities]
+input = ["text"]
+output = ["text"]

models/baidu/ernie-4.5-21b-a3b.toml

@@ -0,0 +1,25 @@
+# https://huggingface.co/baidu/ERNIE-4.5-21B-A3B-PT
+# https://novita.ai/models/model-detail/baidu-ernie-4.5-21B-a3b
+name = "ERNIE 4.5 21B A3B"
+description = "Open ERNIE text MoE with 21B total and 3B active parameters"
+family = "ernie"
+release_date = "2025-06-30"
+last_updated = "2025-06-30"
+attachment = false
+reasoning = false
+tool_call = true
+temperature = true
+open_weights = true
+license = "Apache-2.0"
+
+[limit]
+context = 131_072
+output = 8_000
+
+[modalities]
+input = ["text"]
+output = ["text"]
+
+[[weights]]
+label = "Hugging Face"
+url = "https://huggingface.co/baidu/ERNIE-4.5-21B-A3B-PT"

models/deepseek/deepseek-r1-0528-qwen3-8b.toml

@@ -0,0 +1,24 @@
+# https://huggingface.co/deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
+name = "DeepSeek R1 0528 Qwen3 8B"
+description = "DeepSeek R1 0528 distilled into Qwen3 8B for mathematics, coding, and reasoning"
+family = "deepseek-thinking"
+release_date = "2025-05-29"
+last_updated = "2025-05-29"
+attachment = false
+reasoning = true
+tool_call = false
+temperature = true
+open_weights = true
+license = "MIT"
+
+[limit]
+context = 131_072
+output = 32_000
+
+[modalities]
+input = ["text"]
+output = ["text"]
+
+[[weights]]
+label = "Hugging Face"
+url = "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528-Qwen3-8B"

models/deepseek/deepseek-r1-0528.toml

@@ -0,0 +1,25 @@
+# https://huggingface.co/deepseek-ai/DeepSeek-R1-0528
+name = "DeepSeek R1 0528"
+description = "May 2025 DeepSeek R1 update with improved mathematics, coding, and reasoning"
+family = "deepseek-thinking"
+release_date = "2025-05-28"
+last_updated = "2025-05-28"
+attachment = false
+reasoning = true
+tool_call = true
+structured_output = true
+temperature = true
+open_weights = true
+license = "MIT"
+
+[limit]
+context = 163_840
+output = 32_768
+
+[modalities]
+input = ["text"]
+output = ["text"]
+
+[[weights]]
+label = "Hugging Face"
+url = "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528"

models/gryphe/mythomax-l2-13b.toml

@@ -0,0 +1,22 @@
+# https://huggingface.co/Gryphe/MythoMax-L2-13b
+name = "MythoMax L2 13B"
+description = "Llama 2 merge combining MythoLogic and Huginn for creative writing and roleplay"
+release_date = "2023-08-10"
+last_updated = "2023-08-10"
+attachment = false
+reasoning = false
+tool_call = false
+temperature = true
+open_weights = true
+
+[limit]
+context = 4_096
+output = 3_200
+
+[modalities]
+input = ["text"]
+output = ["text"]
+
+[[weights]]
+label = "Hugging Face"
+url = "https://huggingface.co/Gryphe/MythoMax-L2-13b"

models/inclusionai/ling-3.0-flash-fin.toml

@@ -2,15 +2,17 @@
 # https://novita.ai/models/model-detail/inclusionai-ling-3.0-flash-fin
 # https://openrouter.ai/api/v1/models/inclusionai/ling-3.0-flash-fin:free/endpoints
 # https://vercel.com/changelog/ling-3-0-flash-fin-now-available-on-ai-gateway-for-free
+# MIT weights published: https://huggingface.co/inclusionAI/Ling-3.0-flash-Fin (2026-09-03)
 name = "Ling 3.0 Flash Fin"
 description = "Finance-enhanced model for financial research, multi-step investment workflows, and long-horizon planning and execution"
 family = "ling"
 release_date = "2026-08-27"
-last_updated = "2026-08-27"
+last_updated = "2026-09-03"
 attachment = false
 reasoning = true
 tool_call = true
-open_weights = false
+open_weights = true
+license = "MIT"
 
 [limit]
 context = 262_144
@@ -20,6 +22,10 @@ output = 32_768
 input = ["text"]
 output = ["text"]
 
+[[weights]]
+label = "Hugging Face"
+url = "https://huggingface.co/inclusionAI/Ling-3.0-flash-Fin"
+
 [[benchmarks]]
 name = "FinFIRST"
 score = 52.85