fix: Bedrock: Mantle models template api on ${AWS_REGION}, but aren't served in every region

2026-08-16 10:24:08 +00:00
9 changed files with 75 additions and 12 deletions
@@ -0,0 +1,19 @@
+# bedrock-runtime global CRIS ID (not Mantle). Model card: https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-openai-gpt-56-luna.html
+# Global CRIS pricing (USD per 1M tokens): short ≤272K / long >272K — input / 30m cache write / cache read / output
+base_model = "openai/gpt-5.6-luna"
+reasoning_options = [{ type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }]
+base_model_omit = ["cost.context_over_200k", "experimental.modes.fast"]
+name = "GPT-5.6 Luna (Global)"
+
+[cost]
+input = 0.20
+output = 1.20
+cache_read = 0.02
+cache_write = 0.25
+
+[[cost.tiers]]
+tier = { size = 272_000 }
+input = 0.40
+output = 1.80
+cache_read = 0.04
+cache_write = 0.50
@@ -0,0 +1,19 @@
+# bedrock-runtime global CRIS ID (not Mantle). Model card: https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-openai-gpt-56-sol.html
+# Global CRIS pricing (USD per 1M tokens): short ≤272K / long >272K — input / 30m cache write / cache read / output
+base_model = "openai/gpt-5.6-sol"
+reasoning_options = [{ type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }]
+base_model_omit = ["cost.context_over_200k", "experimental.modes.fast"]
+name = "GPT-5.6 Sol (Global)"
+
+[cost]
+input = 5.00
+output = 30.00
+cache_read = 0.50
+cache_write = 6.25
+
+[[cost.tiers]]
+tier = { size = 272_000 }
+input = 10.00
+output = 45.00
+cache_read = 1.00
+cache_write = 12.50
@@ -0,0 +1,19 @@
+# bedrock-runtime global CRIS ID (not Mantle). Model card: https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-openai-gpt-56-terra.html
+# Global CRIS pricing (USD per 1M tokens): short ≤272K / long >272K — input / 30m cache write / cache read / output
+base_model = "openai/gpt-5.6-terra"
+reasoning_options = [{ type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }]
+base_model_omit = ["cost.context_over_200k", "experimental.modes.fast"]
+name = "GPT-5.6 Terra (Global)"
+
+[cost]
+input = 2.00
+output = 12.00
+cache_read = 0.20
+cache_write = 2.50
+
+[[cost.tiers]]
+tier = { size = 272_000 }
+input = 4.00
+output = 18.00
+cache_read = 0.40
+cache_write = 5.00
@@ -1,3 +1,5 @@
+# Mantle in-region only in us-east-1/us-east-2/us-west-2 (plus gov); pin us-east-1 so non-US AWS_REGION does not 404.
+# Model card: https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-openai-gpt-54.html
 base_model = "openai/gpt-5.4"
 reasoning_options = [{ type = "effort", values = ["none", "low", "medium", "high", "xhigh"] }]
 base_model_omit = ["cost.tiers", "cost.context_over_200k", "experimental.modes.fast", "limit.input"]
@@ -14,5 +16,5 @@ output = 128_000

 [provider]
 npm = "@ai-sdk/amazon-bedrock/mantle"
-api = "https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1"
+api = "https://bedrock-mantle.us-east-1.api.aws/openai/v1"
 shape = "responses"
@@ -1,3 +1,5 @@
+# Mantle in-region only in us-east-1/us-east-2; pin us-east-1 so non-US AWS_REGION does not 404.
+# Model card: https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-openai-gpt-55.html
 base_model = "openai/gpt-5.5"
 reasoning_options = [{ type = "effort", values = ["none", "low", "medium", "high", "xhigh"] }]
 base_model_omit = ["cost.tiers", "cost.context_over_200k", "experimental.modes.fast", "limit.input"]
@@ -14,5 +16,5 @@ output = 128_000

 [provider]
 npm = "@ai-sdk/amazon-bedrock/mantle"
-api = "https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1"
+api = "https://bedrock-mantle.us-east-1.api.aws/openai/v1"
 shape = "responses"
@@ -1,6 +1,6 @@
+# Mantle in-region only in us-east-1/us-east-2/us-west-2; pin us-east-1 so non-US AWS_REGION does not 404.
+# Model card (regions, mantle URL, in-region pricing): https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-openai-gpt-56-luna.html
 # Bedrock model IDs and reasoning support: https://github.com/openai/codex/pull/30285
-# Bedrock context window: https://developers.openai.com/api/docs/guides/amazon-bedrock
-# In-region on-demand pricing (US East N. Virginia & Ohio; also US West Oregon): https://aws.amazon.com/bedrock/pricing/
 # Columns: input / 30m cache write / cache read / output (USD per 1M tokens)
 base_model = "openai/gpt-5.6-luna"
 reasoning_options = [{ type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }]
@@ -21,5 +21,5 @@ cache_write = 0.55

 [provider]
 npm = "@ai-sdk/amazon-bedrock/mantle"
-api = "https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1"
+api = "https://bedrock-mantle.us-east-1.api.aws/openai/v1"
 shape = "responses"
@@ -1,6 +1,6 @@
+# Mantle in-region only in us-east-1/us-east-2; pin us-east-1 so non-US AWS_REGION does not 404.
+# Model card (regions, mantle URL, in-region pricing): https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-openai-gpt-56-sol.html
 # Bedrock model IDs and reasoning support: https://github.com/openai/codex/pull/30285
-# Bedrock context window: https://developers.openai.com/api/docs/guides/amazon-bedrock
-# In-region on-demand pricing (US East N. Virginia & Ohio): https://aws.amazon.com/bedrock/pricing/
 # Columns: input / 30m cache write / cache read / output (USD per 1M tokens)
 base_model = "openai/gpt-5.6-sol"
 reasoning_options = [{ type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }]
@@ -21,5 +21,5 @@ cache_write = 13.75

 [provider]
 npm = "@ai-sdk/amazon-bedrock/mantle"
-api = "https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1"
+api = "https://bedrock-mantle.us-east-1.api.aws/openai/v1"
 shape = "responses"
@@ -1,6 +1,6 @@
+# Mantle in-region only in us-east-1/us-east-2/us-west-2; pin us-east-1 so non-US AWS_REGION does not 404.
+# Model card (regions, mantle URL, in-region pricing): https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-openai-gpt-56-terra.html
 # Bedrock model IDs and reasoning support: https://github.com/openai/codex/pull/30285
-# Bedrock context window: https://developers.openai.com/api/docs/guides/amazon-bedrock
-# In-region on-demand pricing (US East N. Virginia & Ohio; also US West Oregon): https://aws.amazon.com/bedrock/pricing/
 # Columns: input / 30m cache write / cache read / output (USD per 1M tokens)
 base_model = "openai/gpt-5.6-terra"
 reasoning_options = [{ type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }]
@@ -21,5 +21,5 @@ cache_write = 5.50

 [provider]
 npm = "@ai-sdk/amazon-bedrock/mantle"
-api = "https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1"
+api = "https://bedrock-mantle.us-east-1.api.aws/openai/v1"
 shape = "responses"
@@ -1,3 +1,5 @@
+# Mantle in-region only in us-east-1/us-east-2/us-west-2; pin us-east-1 so non-US AWS_REGION does not 404.
+# Model card: https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-xai-grok-4-3.html
 base_model = "xai/grok-4.3"
 reasoning_options = [{ type = "effort", values = ["none", "low", "medium", "high"] }]
 last_updated = "2026-06-28"
@@ -16,5 +18,5 @@ input = ["text", "image"]

 [provider]
 npm = "@ai-sdk/amazon-bedrock/mantle"
-api = "https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1"
+api = "https://bedrock-mantle.us-east-1.api.aws/openai/v1"
 shape = "responses"