chore(sync): update core ZenMux model families

models/deepseek/deepseek-r1-0528.toml

@@ -0,0 +1,29 @@
+# Sources:
+# - https://huggingface.co/deepseek-ai/DeepSeek-R1-0528
+# - https://zenmux.ai/models
+# Accessed 2026-09-19.
+name = "DeepSeek R1 0528"
+description = "May 2025 update to DeepSeek R1 with stronger reasoning, math, coding, and tool use"
+family = "deepseek-thinking"
+release_date = "2025-05-28"
+last_updated = "2025-05-28"
+attachment = false
+reasoning = true
+temperature = true
+tool_call = true
+structured_output = true
+knowledge = "2024-07"
+open_weights = true
+license = "MIT"
+
+[limit]
+context = 128_000
+output = 32_768
+
+[modalities]
+input = ["text"]
+output = ["text"]
+
+[[weights]]
+label = "Hugging Face"
+url = "https://huggingface.co/deepseek-ai/DeepSeek-R1-0528"

models/deepseek/deepseek-v3.2-exp.toml

@@ -0,0 +1,30 @@
+# Sources:
+# - https://zenmux.ai/models
+# - https://zenmux.ai/docs/api/openai/openai-list-models.html
+# - https://zenmux.ai/docs/api/anthropic/anthropic-list-models.html
+# - https://zenmux.ai/docs/api/vertexai/google-list-models.html
+# - https://huggingface.co/deepseek-ai/DeepSeek-V3.2-Exp
+# Accessed 2026-09-19.
+name = "DeepSeek-V3.2-Exp"
+description = "DeepSeek-V3.2-Exp is an experimental model version, serving as an intermediate step toward the next-generation architecture. Built on the foundation of V3.1-Terminus, it introduces the DeepSeek Sparse Attention (DSA) mechanism—a sparse attention mechanism designed to explore and validate the optimization of training and inference efficiency in long-context scenarios. This experimental version represents the team's continuous research on more efficient Transformer architectures, with a specific focus on improving computational efficiency when processing long text sequences. For the first time, DSA enables fine-grained sparse attention, which significantly enhances the efficiency of long-context training and inference while maintaining almost unchanged model output quality."
+family = "deepseek"
+release_date = "2025-09-29"
+last_updated = "2025-09-29"
+attachment = false
+reasoning = true
+temperature = true
+tool_call = true
+structured_output = true
+open_weights = true
+
+[limit]
+context = 163_840
+output = 65_536
+
+[modalities]
+input = ["text"]
+output = ["text"]
+
+[[weights]]
+label = "Hugging Face"
+url = "https://huggingface.co/deepseek-ai/DeepSeek-V3.2-Exp"

models/google/gemini-omni-1.1-flash-preview.toml

@@ -0,0 +1,25 @@
+# Sources:
+# - https://zenmux.ai/models
+# - https://zenmux.ai/docs/api/openai/openai-list-models.html
+# - https://zenmux.ai/docs/api/anthropic/anthropic-list-models.html
+# - https://zenmux.ai/docs/api/vertexai/google-list-models.html
+# Accessed 2026-09-19.
+name = "Gemini Omni 1.1 Flash Preview"
+description = "Gemini Omni 1.1 Flash (Preview) is a multimodal model designed for video, image, and text tasks. It is optimized for video generation, offering video output alongside text responses in a single model."
+family = "gemini"
+release_date = "2026-08-31"
+last_updated = "2026-08-31"
+attachment = true
+reasoning = true
+temperature = false
+tool_call = false
+structured_output = false
+open_weights = false
+
+[limit]
+context = 131_072
+output = 57_920
+
+[modalities]
+input = ["text", "image", "video"]
+output = ["text", "video"]

models/google/veo-3.1-fast-generate-001.toml

@@ -0,0 +1,25 @@
+# Sources:
+# - https://zenmux.ai/models
+# - https://zenmux.ai/docs/api/openai/openai-list-models.html
+# - https://zenmux.ai/docs/api/anthropic/anthropic-list-models.html
+# - https://zenmux.ai/docs/api/vertexai/google-list-models.html
+# Accessed 2026-09-19.
+name = "Veo 3.1 Fast"
+description = "Veo 3.1 Fast is a speed-optimized variant of Google DeepMind's flagship video generation model. It is designed to generate high-quality video significantly faster and at a lower cost than the standard Veo 3.1 Quality model, making it ideal for rapid prototyping and high-volume content creation."
+family = "veo"
+release_date = "2025-10-15"
+last_updated = "2026-01-01"
+attachment = true
+reasoning = false
+temperature = false
+tool_call = false
+structured_output = false
+open_weights = false
+
+[limit]
+context = 1_024
+output = 0
+
+[modalities]
+input = ["text", "image", "video"]
+output = ["video"]

models/google/veo-3.1-generate-001.toml

@@ -0,0 +1,25 @@
+# Sources:
+# - https://zenmux.ai/models
+# - https://zenmux.ai/docs/api/openai/openai-list-models.html
+# - https://zenmux.ai/docs/api/anthropic/anthropic-list-models.html
+# - https://zenmux.ai/docs/api/vertexai/google-list-models.html
+# Accessed 2026-09-19.
+name = "Veo 3.1"
+description = "Veo 3.1 is Google's state-of-the-art model for generating high-fidelity, 8-second 720p, 1080p or 4k videos featuring stunning realism and natively generated audio. You can access this model programmatically using the Gemini API. To learn more about the available Veo model variants, see the Model Versions section."
+family = "veo"
+release_date = "2025-10-15"
+last_updated = "2026-01"
+attachment = true
+reasoning = false
+temperature = false
+tool_call = false
+structured_output = false
+open_weights = false
+
+[limit]
+context = 1_024
+output = 0
+
+[modalities]
+input = ["text", "image"]
+output = ["video"]

models/google/veo-3.1-lite-generate-001.toml

@@ -0,0 +1,25 @@
+# Sources:
+# - https://zenmux.ai/models
+# - https://zenmux.ai/docs/api/openai/openai-list-models.html
+# - https://zenmux.ai/docs/api/anthropic/anthropic-list-models.html
+# - https://zenmux.ai/docs/api/vertexai/google-list-models.html
+# Accessed 2026-09-19.
+name = "Veo 3.1 Lite"
+description = "Veo 3.1 Lite is Google DeepMind's most cost-efficient AI video generation model, released on March 31, 2026. It is designed to provide professional-grade video capabilities at a significantly lower price point, making it ideal for developers and content teams who need to scale high-volume video production"
+family = "veo"
+release_date = "2026-03-31"
+last_updated = "2026-03-31"
+attachment = true
+reasoning = false
+temperature = false
+tool_call = false
+structured_output = false
+open_weights = false
+
+[limit]
+context = 1_024
+output = 0
+
+[modalities]
+input = ["text", "image"]
+output = ["video"]

models/openai/gpt-transcribe.toml

@@ -0,0 +1,25 @@
+# Sources:
+# - https://zenmux.ai/models
+# - https://zenmux.ai/docs/api/openai/openai-list-models.html
+# - https://zenmux.ai/docs/api/anthropic/anthropic-list-models.html
+# - https://zenmux.ai/docs/api/vertexai/google-list-models.html
+# Accessed 2026-09-19.
+name = "GPT Transcribe"
+description = "GPT Transcribe is a high-accuracy speech-to-text model from OpenAI. It is suited for recorded audio, streamed file transcription, and committed Realtime turns, with free-form context, keyword hints, and multiple language hints for specialized terms and multilingual speech."
+family = "gpt"
+release_date = "2026-08-06"
+last_updated = "2026-08-06"
+attachment = true
+reasoning = false
+temperature = false
+tool_call = false
+structured_output = false
+open_weights = false
+
+[limit]
+context = 15_000
+output = 15_000
+
+[modalities]
+input = ["audio"]
+output = ["text"]

models/openai/text-embedding-3-large.toml

@@ -0,0 +1,25 @@
+# Sources:
+# - https://zenmux.ai/models
+# - https://zenmux.ai/docs/api/openai/openai-list-models.html
+# - https://zenmux.ai/docs/api/anthropic/anthropic-list-models.html
+# - https://zenmux.ai/docs/api/vertexai/google-list-models.html
+# Accessed 2026-09-19.
+name = "text-embedding-3-large"
+description = "Embedding model for semantic search, retrieval, clustering, and ranking pipelines"
+family = "text-embedding"
+release_date = "2024-01-25"
+last_updated = "2024-01-25"
+attachment = false
+reasoning = false
+temperature = false
+tool_call = false
+knowledge = "2024-01"
+open_weights = false
+
+[limit]
+context = 8_191
+output = 3_072
+
+[modalities]
+input = ["text"]
+output = ["text"]

models/openai/text-embedding-3-small.toml

@@ -0,0 +1,25 @@
+# Sources:
+# - https://zenmux.ai/models
+# - https://zenmux.ai/docs/api/openai/openai-list-models.html
+# - https://zenmux.ai/docs/api/anthropic/anthropic-list-models.html
+# - https://zenmux.ai/docs/api/vertexai/google-list-models.html
+# Accessed 2026-09-19.
+name = "text-embedding-3-small"
+description = "Embedding model for semantic search, retrieval, clustering, and ranking pipelines"
+family = "text-embedding"
+release_date = "2024-01-25"
+last_updated = "2024-01-25"
+attachment = false
+reasoning = false
+temperature = false
+tool_call = false
+knowledge = "2024-01"
+open_weights = false
+
+[limit]
+context = 8_191
+output = 1_536
+
+[modalities]
+input = ["text"]
+output = ["text"]

providers/zenmux/models/anthropic/claude-fable-5.1.toml

@@ -0,0 +1,17 @@
+# Effort: reasoning_effort = low|medium|high|xhigh|max
+# https://zenmux.ai/docs/guide/advanced/reasoning.html
+base_model = "anthropic/claude-fable-5-1"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "xhigh", "max"]
+
+[cost]
+input = 10
+output = 50
+cache_read = 0.25
+cache_write = 12.5
+
+[provider]
+npm = "@ai-sdk/anthropic"
+api = "https://zenmux.ai/api/anthropic/v1"

providers/zenmux/models/anthropic/claude-fable-5.toml

@@ -1,11 +1,16 @@
+# Effort: reasoning_effort = low|medium|high|xhigh|max
+# https://zenmux.ai/docs/guide/advanced/reasoning.html
 base_model = "anthropic/claude-fable-5"
-reasoning_options = [{ type = "effort", values = ["low", "medium", "high"] }]
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high", "xhigh", "max"]
 
 [cost]
-input = 10.00
-output = 50.00
-cache_read = 1.00
-cache_write = 12.50
+input = 10
+output = 50
+cache_read = 1
+cache_write = 12.5
 
 [provider]
 npm = "@ai-sdk/anthropic"

providers/zenmux/models/anthropic/claude-haiku-4.5.toml

@@ -1,28 +1,19 @@
+# Budget: thinking.budget_tokens (integer)
+# https://zenmux.ai/docs/guide/advanced/reasoning.html
+base_model = "anthropic/claude-haiku-4-5"
 name = "Claude Haiku 4.5"
-description = "Fast Claude model for responsive assistance, classification, and lightweight agents"
-release_date = "2025-10-15"
-last_updated = "2025-10-15"
-attachment = true
-reasoning = false
-temperature = true
-tool_call = true
-knowledge = "2025-01-01"
-open_weights = false
+structured_output = true
+
+[[reasoning_options]]
+type = "budget_tokens"
+min = 1_024
 
 [cost]
-input = 1.00
-output = 5.00
-cache_read = 0.10
+input = 1
+output = 5
+cache_read = 0.1
 cache_write = 1.25
 
-[limit]
-context = 200_000
-output = 64_000
-
-[modalities]
-input = ["image", "text"]
-output = ["text"]
-
 [provider]
 npm = "@ai-sdk/anthropic"
 api = "https://zenmux.ai/api/anthropic/v1"