fix: clean up OpenRouter-generated provider configs (#37)

- Merge unique models from duplicate providers into their hand-written
  counterparts and remove the duplicates:
  - alibaba (tongyi-deepresearch) → qwen
  - amazon (nova-2-lite, nova-micro, nova-premier) → bedrock
  - bytedance (ui-tars) → volcengine
  - nvidia (nemotron-3-nano, nemotron-3-super, etc.) → nvidia-nim
  - rekaai (reka-flash-3) → reka
- Set correct official API base_url for providers with public APIs:
  arcee-ai, inception, morph, reka, upstage
- Set key_required=false for 20 providers only accessible through
  hosting platforms (no public API)
- Update sync-pricing.py with SKIP_DUPLICATES, PROVIDER_API mapping,
  and default key_required=false for future auto-generated providers

providers/aion-labs.toml

@@ -5,7 +5,7 @@ id = "aion-labs"
 display_name = "Aion Labs"
 api_key_env = "AION_LABS_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "aion-1.0"

providers/alibaba.toml

@@ -1,18 +0,0 @@
-# alibaba — auto-generated from OpenRouter API
-
-[provider]
-id = "alibaba"
-display_name = "Alibaba"
-api_key_env = "ALIBABA_API_KEY"
-base_url = ""
-key_required = true
-
-[[models]]
-id = "tongyi-deepresearch-30b-a3b"
-display_name = "Tongyi DeepResearch 30B A3B"
-tier = "fast"
-context_window = 131072
-max_output_tokens = 131072
-input_cost_per_m = 0.09
-output_cost_per_m = 0.45
-supports_streaming = true

providers/allenai.toml

@@ -5,7 +5,7 @@ id = "allenai"
 display_name = "Allenai"
 api_key_env = "ALLENAI_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "olmo-2-0325-32b-instruct"

providers/amazon.toml

@@ -1,58 +0,0 @@
-# amazon — auto-generated from OpenRouter API
-
-[provider]
-id = "amazon"
-display_name = "Amazon"
-api_key_env = "AMAZON_API_KEY"
-base_url = ""
-key_required = true
-
-[[models]]
-id = "nova-2-lite-v1"
-display_name = "Amazon: Nova 2 Lite"
-tier = "fast"
-context_window = 1000000
-max_output_tokens = 65535
-input_cost_per_m = 0.3
-output_cost_per_m = 2.5
-supports_streaming = true
-
-[[models]]
-id = "nova-lite-v1"
-display_name = "Amazon: Nova Lite 1.0"
-tier = "fast"
-context_window = 300000
-max_output_tokens = 5120
-input_cost_per_m = 0.06
-output_cost_per_m = 0.24
-supports_streaming = true
-
-[[models]]
-id = "nova-micro-v1"
-display_name = "Amazon: Nova Micro 1.0"
-tier = "fast"
-context_window = 128000
-max_output_tokens = 5120
-input_cost_per_m = 0.035
-output_cost_per_m = 0.14
-supports_streaming = true
-
-[[models]]
-id = "nova-premier-v1"
-display_name = "Amazon: Nova Premier 1.0"
-tier = "smart"
-context_window = 1000000
-max_output_tokens = 32000
-input_cost_per_m = 2.5
-output_cost_per_m = 12.5
-supports_streaming = true
-
-[[models]]
-id = "nova-pro-v1"
-display_name = "Amazon: Nova Pro 1.0"
-tier = "smart"
-context_window = 300000
-max_output_tokens = 5120
-input_cost_per_m = 0.8
-output_cost_per_m = 3.2
-supports_streaming = true

providers/arcee-ai.toml

@@ -3,8 +3,8 @@
 [provider]
 id = "arcee-ai"
 display_name = "Arcee Ai"
-api_key_env = "ARCEE_AI_API_KEY"
+api_key_env = "ARCEE_API_KEY"
-base_url = ""
+base_url = "https://api.arcee.ai/v1"
 key_required = true
 
 [[models]]

providers/bedrock.toml

@@ -111,3 +111,33 @@ supports_tools = true
 supports_vision = false
 supports_streaming = true
 aliases = []
+
+[[models]]
+id = "bedrock/amazon.nova-2-lite-v1:0"
+display_name = "Amazon: Nova 2 Lite"
+tier = "fast"
+context_window = 1000000
+max_output_tokens = 65535
+input_cost_per_m = 0.3
+output_cost_per_m = 2.5
+supports_streaming = true
+
+[[models]]
+id = "bedrock/amazon.nova-micro-v1:0"
+display_name = "Amazon: Nova Micro 1.0"
+tier = "fast"
+context_window = 128000
+max_output_tokens = 5120
+input_cost_per_m = 0.035
+output_cost_per_m = 0.14
+supports_streaming = true
+
+[[models]]
+id = "bedrock/amazon.nova-premier-v1:0"
+display_name = "Amazon: Nova Premier 1.0"
+tier = "frontier"
+context_window = 1000000
+max_output_tokens = 65535
+input_cost_per_m = 2.5
+output_cost_per_m = 12.5
+supports_streaming = true

providers/bytedance.toml

@@ -1,18 +0,0 @@
-# bytedance — auto-generated from OpenRouter API
-
-[provider]
-id = "bytedance"
-display_name = "Bytedance"
-api_key_env = "BYTEDANCE_API_KEY"
-base_url = ""
-key_required = true
-
-[[models]]
-id = "ui-tars-1.5-7b"
-display_name = "ByteDance: UI-TARS 7B "
-tier = "fast"
-context_window = 128000
-max_output_tokens = 2048
-input_cost_per_m = 0.1
-output_cost_per_m = 0.2
-supports_streaming = true

providers/deepcogito.toml

@@ -5,7 +5,7 @@ id = "deepcogito"
 display_name = "Deepcogito"
 api_key_env = "DEEPCOGITO_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "cogito-v2.1-671b"

providers/eleutherai.toml

@@ -5,7 +5,7 @@ id = "eleutherai"
 display_name = "Eleutherai"
 api_key_env = "ELEUTHERAI_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "llemma_7b"

providers/essentialai.toml

@@ -5,7 +5,7 @@ id = "essentialai"
 display_name = "Essentialai"
 api_key_env = "ESSENTIALAI_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "rnj-1-instruct"

providers/ibm-granite.toml

@@ -5,7 +5,7 @@ id = "ibm-granite"
 display_name = "Ibm Granite"
 api_key_env = "IBM_GRANITE_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "granite-4.0-h-micro"

providers/inception.toml

@@ -4,7 +4,7 @@
 id = "inception"
 display_name = "Inception"
 api_key_env = "INCEPTION_API_KEY"
-base_url = ""
+base_url = "https://api.inceptionlabs.ai/v1"
 key_required = true
 
 [[models]]

providers/inflection.toml

@@ -5,7 +5,7 @@ id = "inflection"
 display_name = "Inflection"
 api_key_env = "INFLECTION_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "inflection-3-pi"

providers/kwaipilot.toml

@@ -5,7 +5,7 @@ id = "kwaipilot"
 display_name = "Kwaipilot"
 api_key_env = "KWAIPILOT_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "kat-coder-pro"

providers/liquid.toml

@@ -5,7 +5,7 @@ id = "liquid"
 display_name = "Liquid"
 api_key_env = "LIQUID_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "lfm-2-24b-a2b"

providers/meituan.toml

@@ -5,7 +5,7 @@ id = "meituan"
 display_name = "Meituan"
 api_key_env = "MEITUAN_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "longcat-flash-chat"

providers/meta-llama.toml

@@ -5,7 +5,7 @@ id = "meta-llama"
 display_name = "Meta Llama"
 api_key_env = "META_LLAMA_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "llama-3-70b-instruct"

providers/microsoft.toml

@@ -5,7 +5,7 @@ id = "microsoft"
 display_name = "Microsoft"
 api_key_env = "MICROSOFT_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "phi-4"

providers/morph.toml

@@ -4,7 +4,7 @@
 id = "morph"
 display_name = "Morph"
 api_key_env = "MORPH_API_KEY"
-base_url = ""
+base_url = "https://api.morphlabs.io/v1"
 key_required = true
 
 [[models]]

providers/nex-agi.toml

@@ -5,7 +5,7 @@ id = "nex-agi"
 display_name = "Nex Agi"
 api_key_env = "NEX_AGI_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "deepseek-v3.1-nex-n1"

providers/nousresearch.toml

@@ -5,7 +5,7 @@ id = "nousresearch"
 display_name = "Nousresearch"
 api_key_env = "NOUSRESEARCH_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "hermes-2-pro-llama-3-8b"

providers/nvidia-nim.toml

@@ -307,3 +307,44 @@ supports_tools = true
 supports_vision = false
 supports_streaming = true
 aliases = ["phi-4-mini"]
+
+[[models]]
+id = "nvidia/nemotron-3-nano-30b-a3b"
+display_name = "Nvidia: Nemotron 3 Nano 30B"
+tier = "fast"
+context_window = 131072
+max_output_tokens = 32768
+input_cost_per_m = 0.0
+output_cost_per_m = 0.0
+supports_streaming = true
+
+[[models]]
+id = "nvidia/nemotron-3-super-120b-a12b"
+display_name = "Nvidia: Nemotron 3 Super 120B"
+tier = "smart"
+context_window = 131072
+max_output_tokens = 32768
+input_cost_per_m = 0.0
+output_cost_per_m = 0.0
+supports_streaming = true
+
+[[models]]
+id = "nvidia/nemotron-nano-12b-v2-vl"
+display_name = "Nvidia: Nemotron Nano 12B V2 VL"
+tier = "fast"
+context_window = 131072
+max_output_tokens = 32768
+input_cost_per_m = 0.0
+output_cost_per_m = 0.0
+supports_vision = true
+supports_streaming = true
+
+[[models]]
+id = "nvidia/nemotron-nano-9b-v2"
+display_name = "Nvidia: Nemotron Nano 9B V2"
+tier = "fast"
+context_window = 131072
+max_output_tokens = 32768
+input_cost_per_m = 0.0
+output_cost_per_m = 0.0
+supports_streaming = true

providers/nvidia.toml

@@ -1,118 +0,0 @@
-# nvidia — auto-generated from OpenRouter API
-
-[provider]
-id = "nvidia"
-display_name = "Nvidia"
-api_key_env = "NVIDIA_API_KEY"
-base_url = ""
-key_required = true
-
-[[models]]
-id = "llama-3.1-nemotron-70b-instruct"
-display_name = "NVIDIA: Llama 3.1 Nemotron 70B Instruct"
-tier = "smart"
-context_window = 131072
-max_output_tokens = 16384
-input_cost_per_m = 1.2
-output_cost_per_m = 1.2
-supports_streaming = true
-
-[[models]]
-id = "llama-3.1-nemotron-ultra-253b-v1"
-display_name = "NVIDIA: Llama 3.1 Nemotron Ultra 253B v1"
-tier = "smart"
-context_window = 131072
-max_output_tokens = 16384
-input_cost_per_m = 0.6
-output_cost_per_m = 1.8
-supports_streaming = true
-
-[[models]]
-id = "llama-3.3-nemotron-super-49b-v1.5"
-display_name = "NVIDIA: Llama 3.3 Nemotron Super 49B V1.5"
-tier = "fast"
-context_window = 131072
-max_output_tokens = 16384
-input_cost_per_m = 0.1
-output_cost_per_m = 0.4
-supports_streaming = true
-
-[[models]]
-id = "nemotron-3-nano-30b-a3b"
-display_name = "NVIDIA: Nemotron 3 Nano 30B A3B"
-tier = "fast"
-context_window = 262144
-max_output_tokens = 16384
-input_cost_per_m = 0.05
-output_cost_per_m = 0.2
-supports_streaming = true
-
-[[models]]
-id = "nemotron-3-nano-30b-a3b:free"
-display_name = "NVIDIA: Nemotron 3 Nano 30B A3B (free)"
-tier = "fast"
-context_window = 256000
-max_output_tokens = 16384
-input_cost_per_m = 0.0
-output_cost_per_m = 0.0
-supports_streaming = true
-
-[[models]]
-id = "nemotron-3-super-120b-a12b"
-display_name = "NVIDIA: Nemotron 3 Super"
-tier = "fast"
-context_window = 262144
-max_output_tokens = 16384
-input_cost_per_m = 0.1
-output_cost_per_m = 0.5
-supports_streaming = true
-
-[[models]]
-id = "nemotron-3-super-120b-a12b:free"
-display_name = "NVIDIA: Nemotron 3 Super (free)"
-tier = "fast"
-context_window = 262144
-max_output_tokens = 262144
-input_cost_per_m = 0.0
-output_cost_per_m = 0.0
-supports_streaming = true
-
-[[models]]
-id = "nemotron-nano-12b-v2-vl"
-display_name = "NVIDIA: Nemotron Nano 12B 2 VL"
-tier = "fast"
-context_window = 131072
-max_output_tokens = 16384
-input_cost_per_m = 0.2
-output_cost_per_m = 0.6
-supports_streaming = true
-
-[[models]]
-id = "nemotron-nano-12b-v2-vl:free"
-display_name = "NVIDIA: Nemotron Nano 12B 2 VL (free)"
-tier = "fast"
-context_window = 128000
-max_output_tokens = 128000
-input_cost_per_m = 0.0
-output_cost_per_m = 0.0
-supports_streaming = true
-
-[[models]]
-id = "nemotron-nano-9b-v2"
-display_name = "NVIDIA: Nemotron Nano 9B V2"
-tier = "fast"
-context_window = 131072
-max_output_tokens = 16384
-input_cost_per_m = 0.04
-output_cost_per_m = 0.16
-supports_streaming = true
-
-[[models]]
-id = "nemotron-nano-9b-v2:free"
-display_name = "NVIDIA: Nemotron Nano 9B V2 (free)"
-tier = "fast"
-context_window = 128000
-max_output_tokens = 16384
-input_cost_per_m = 0.0
-output_cost_per_m = 0.0
-supports_streaming = true

providers/prime-intellect.toml

@@ -5,7 +5,7 @@ id = "prime-intellect"
 display_name = "Prime Intellect"
 api_key_env = "PRIME_INTELLECT_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "intellect-3"

providers/qwen.toml

@@ -157,3 +157,13 @@ supports_tools = false
 supports_vision = true
 supports_streaming = true
 aliases = []
+
+[[models]]
+id = "tongyi-deepresearch-30b-a3b"
+display_name = "Tongyi DeepResearch 30B A3B"
+tier = "fast"
+context_window = 131072
+max_output_tokens = 131072
+input_cost_per_m = 0.09
+output_cost_per_m = 0.45
+supports_streaming = true

providers/reka.toml

@@ -4,7 +4,7 @@
 id = "reka"
 display_name = "Reka"
 api_key_env = "REKA_API_KEY"
-base_url = ""
+base_url = "https://api.reka.ai/v1"
 key_required = true
 
 [[models]]
@@ -16,3 +16,13 @@ max_output_tokens = 16384
 input_cost_per_m = 0.1
 output_cost_per_m = 0.1
 supports_streaming = true
+
+[[models]]
+id = "reka-flash-3"
+display_name = "Reka: Flash 3"
+tier = "fast"
+context_window = 131072
+max_output_tokens = 32768
+input_cost_per_m = 0.15
+output_cost_per_m = 0.6
+supports_streaming = true

providers/rekaai.toml

@@ -1,18 +0,0 @@
-# rekaai — auto-generated from OpenRouter API
-
-[provider]
-id = "rekaai"
-display_name = "Rekaai"
-api_key_env = "REKAAI_API_KEY"
-base_url = ""
-key_required = true
-
-[[models]]
-id = "reka-flash-3"
-display_name = "Reka: Flash 3"
-tier = "fast"
-context_window = 65536
-max_output_tokens = 65536
-input_cost_per_m = 0.1
-output_cost_per_m = 0.2
-supports_streaming = true

providers/relace.toml

@@ -5,7 +5,7 @@ id = "relace"
 display_name = "Relace"
 api_key_env = "RELACE_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "relace-apply-3"

providers/switchpoint.toml

@@ -5,7 +5,7 @@ id = "switchpoint"
 display_name = "Switchpoint"
 api_key_env = "SWITCHPOINT_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "router"

providers/tencent.toml

@@ -5,7 +5,7 @@ id = "tencent"
 display_name = "Tencent"
 api_key_env = "TENCENT_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "hunyuan-a13b-instruct"

providers/tngtech.toml

@@ -5,7 +5,7 @@ id = "tngtech"
 display_name = "Tngtech"
 api_key_env = "TNGTECH_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "deepseek-r1t2-chimera"

providers/upstage.toml

@@ -4,7 +4,7 @@
 id = "upstage"
 display_name = "Upstage"
 api_key_env = "UPSTAGE_API_KEY"
-base_url = ""
+base_url = "https://api.upstage.ai/v1"
 key_required = true
 
 [[models]]

providers/volcengine.toml

@@ -98,3 +98,13 @@ supports_tools = true
 supports_vision = true
 supports_streaming = true
 aliases = ["doubao-vision"]
+
+[[models]]
+id = "ui-tars-1.5-7b"
+display_name = "ByteDance: UI-TARS 7B"
+tier = "fast"
+context_window = 128000
+max_output_tokens = 2048
+input_cost_per_m = 0.1
+output_cost_per_m = 0.2
+supports_streaming = true

providers/xiaomi.toml

@@ -5,7 +5,7 @@ id = "xiaomi"
 display_name = "Xiaomi"
 api_key_env = "XIAOMI_API_KEY"
 base_url = ""
-key_required = true
+key_required = false
 
 [[models]]
 id = "mimo-v2-flash"

scripts/sync-pricing.py

@@ -36,6 +36,27 @@ SKIP_PROVIDERS = {
     "anthracite-org", "alpindale", "alfredpros", "mancer",
 }
 
+# Skip creating NEW provider files for these — they overlap with hand-written
+# providers but may contain unique models, so existing files are kept.
+SKIP_DUPLICATES = {
+    "alibaba",      # overlaps with qwen.toml but has unique models
+    "amazon",       # overlaps with bedrock.toml but has unique models
+    "bytedance",    # overlaps with volcengine.toml but has unique models
+    "nvidia",       # overlaps with nvidia-nim.toml but has unique models
+    "rekaai",       # overlaps with reka.toml but has unique models
+}
+
+# Providers with known public APIs — set their official base_url + api_key_env.
+# Providers NOT in this map are OpenRouter-only and get key_required = false.
+PROVIDER_API = {
+    "arcee-ai":     ("https://api.arcee.ai/v1", "ARCEE_API_KEY"),
+    "inception":    ("https://api.inceptionlabs.ai/v1", "INCEPTION_API_KEY"),
+    "morph":        ("https://api.morphlabs.io/v1", "MORPH_API_KEY"),
+    "nvidia":       ("https://integrate.api.nvidia.com/v1", "NVIDIA_API_KEY"),
+    "reka":         ("https://api.reka.ai/v1", "REKA_API_KEY"),
+    "upstage":      ("https://api.upstage.ai/v1", "UPSTAGE_API_KEY"),
+}
+
 
 def fetch_openrouter_models():
     """Fetch all models from OpenRouter."""
@@ -119,8 +140,18 @@ def generate_provider_toml(provider_id, models, dry_run=False):
     if toml_path.exists():
         return 0
 
-    # Derive api_key_env from provider name: FOO_BAR -> FOO_BAR_API_KEY
+    if our_name in SKIP_DUPLICATES:
-    env_key = our_name.upper().replace("-", "_") + "_API_KEY"
+        return 0
+
+    # Check if provider has a known public API
+    if our_name in PROVIDER_API:
+        base_url, env_key = PROVIDER_API[our_name]
+        key_required = "true"
+    else:
+        # No known public API — only accessible via OpenRouter/hosting providers
+        env_key = our_name.upper().replace("-", "_") + "_API_KEY"
+        base_url = ""
+        key_required = "false"
 
     lines = [
         f'# {provider_id} — auto-generated from OpenRouter API',
@@ -129,8 +160,8 @@ def generate_provider_toml(provider_id, models, dry_run=False):
         f'id = "{our_name}"',
         f'display_name = "{provider_id.replace("-", " ").title()}"',
         f'api_key_env = "{env_key}"',
-        f'base_url = ""',
+        f'base_url = "{base_url}"',
-        f"key_required = true",
+        f"key_required = {key_required}",
         f"",
     ]