Sync NVIDIA endpoint model catalog

providers/nvidia/models/abacusai/dracarys-llama-3_1-70b-instruct.toml

@@ -0,0 +1,20 @@
+name = "dracarys-llama-3.1-70b-instruct"
+release_date = "2024-09-11"
+last_updated = "2025-05-22"
+attachment = false
+reasoning = false
+temperature = true
+tool_call = true
+open_weights = true
+
+[cost]
+input = 0.0
+output = 0.0
+
+[limit]
+context = 128_000
+output = 8_192
+
+[modalities]
+input = ["text"]
+output = ["text"]

providers/nvidia/models/baai/bge-m3.toml

@@ -0,0 +1,21 @@
+name = "BGE M3"
+family = "bge"
+release_date = "2024-01-30"
+last_updated = "2026-04-30"
+attachment = false
+reasoning = false
+temperature = false
+tool_call = false
+open_weights = true
+
+[cost]
+input = 0.0
+output = 0.0
+
+[limit]
+context = 8_192
+output = 1_024
+
+[modalities]
+input = ["text"]
+output = ["text"]

providers/nvidia/models/black-forest-labs/flux_1-kontext-dev.toml

@@ -0,0 +1,20 @@
+name = "FLUX.1-Kontext-dev"
+release_date = "2025-08-12"
+last_updated = "2025-08-12"
+attachment = true
+reasoning = false
+temperature = false
+tool_call = false
+open_weights = true
+
+[cost]
+input = 0.0
+output = 0.0
+
+[limit]
+context = 40_960
+output = 40_960
+
+[modalities]
+input = ["text", "image"]
+output = ["image"]

providers/nvidia/models/black-forest-labs/flux_1-schnell.toml

@@ -0,0 +1,23 @@
+name = "FLUX.1-schnell"
+release_date = "2024-08-01"
+last_updated = "2026-02-04"
+attachment = false
+reasoning = false
+temperature = false
+knowledge = "2024-07"
+tool_call = false
+structured_output = false
+open_weights = true
+
+[cost]
+input = 0.0
+output = 0.0
+
+[limit]
+context = 77
+input = 77
+output = 0
+
+[modalities]
+input = ["text"]
+output = ["image"]

providers/nvidia/models/black-forest-labs/flux_2-klein-4b.toml

@@ -0,0 +1,22 @@
+name = "FLUX.2 Klein 4B"
+family = "flux"
+release_date = "2026-01-14"
+last_updated = "2026-01-31"
+attachment = false
+reasoning = false
+temperature = true
+knowledge = "2025-06"
+tool_call = false
+open_weights = true
+
+[cost]
+input = 0.0
+output = 0.0
+
+[limit]
+context = 40_960
+output = 40_960
+
+[modalities]
+input = ["image", "text"]
+output = ["image"]

providers/nvidia/models/nvidia/llama-3.1-nemotron-51b-instruct.toml → providers/nvidia/models/bytedance/seed-oss-36b-instruct.toml

@@ -1,20 +1,21 @@
-name = "Llama 3.1 Nemotron 51b Instruct"
+name = "ByteDance-Seed/Seed-OSS-36B-Instruct"
+family = "seed"
+release_date = "2025-09-04"
+last_updated = "2025-11-25"
 attachment = false
 reasoning = false
 temperature = true
 tool_call = true
 structured_output = true
-release_date = "2024-09-22"
-last_updated = "2024-09-22"
 open_weights = false
 
 [cost]
-input = 0.00
-output = 0.00
+input = 0.0
+output = 0.0
 
 [limit]
-context = 128000
-output = 4096
+context = 262_000
+output = 262_000
 
 [modalities]
 input = ["text"]

providers/nvidia/models/google/google-paligemma.toml

@@ -0,0 +1,20 @@
+name = "paligemma"
+release_date = "2024-05-14"
+last_updated = "2024-08-26"
+attachment = true
+reasoning = false
+temperature = true
+tool_call = false
+open_weights = true
+
+[cost]
+input = 0.0
+output = 0.0
+
+[limit]
+context = 128_000
+output = 8_192
+
+[modalities]
+input = ["text", "image"]
+output = ["text"]