[BOT ISSUE] Missing AWS Bedrock models: openai.gpt-5.5, openai.gpt-5.4

[github-actions]

Gap

Amazon Bedrock now offers OpenAI GPT-5.5 and GPT-5.4 as managed models. Both are listed on the Bedrock supported models page and have individual model cards confirming the model IDs openai.gpt-5.5 and openai.gpt-5.4.

Neither entry exists in model_list.json. The base gpt-5.5 and gpt-5.4 entries have available_providers: ["openai", "azure"] — Bedrock is absent.

These models use the Bedrock Responses API endpoint (bedrock-mantle), not the standard Converse API. Existing Bedrock OpenAI entries (openai.gpt-oss-*) use format: "converse", but GPT-5.5/5.4 go through the OpenAI-compatible endpoint and may need format: "openai".

Models

Model ID	Bedrock Input $/M	Bedrock Output $/M	Cached Input $/M	Context	Regions
openai.gpt-5.5	$5.50	$33.00	$0.55	272K	us-east-1, us-east-2
openai.gpt-5.4	$2.75	$16.50	$0.275	272K	us-east-1, us-east-2, us-west-2, us-gov-west-1

Note: Bedrock pricing is ~10% above OpenAI direct pricing ($5.00/$30.00 for GPT-5.5, $2.50/$15.00 for GPT-5.4), so these need separate Bedrock-specific entries rather than simply adding "bedrock" to the base models' available_providers.

Sources

• AWS Bedrock supported models — lists both under OpenAI
• AWS Bedrock pricing — per-token costs confirmed
• Individual model cards: GPT-5.5, GPT-5.4 — model IDs confirmed in code samples

Verification

• Cross-source confirmation: model IDs appear on both the supported models list and individual model cards with programmatic access samples
• Recent commits: no prior additions of openai.gpt-5.5 or openai.gpt-5.4 in git history
• ID format: matches existing Bedrock OpenAI convention (openai.{model}) per openai.gpt-oss-120b-1:0, openai.gpt-oss-20b-1:0

{
  "kind": "missing_model",
  "provider": "bedrock",
  "models": ["openai.gpt-5.5", "openai.gpt-5.4"],
  "status": "active",
  "model_specs": {
    "openai.gpt-5.5": {
      "format": "openai",
      "flavor": "chat",
      "multimodal": true,
      "input_cost_per_mil_tokens": 5.5,
      "output_cost_per_mil_tokens": 33,
      "input_cache_read_cost_per_mil_tokens": 0.55,
      "displayName": "OpenAI GPT-5.5 (Bedrock)",
      "reasoning": true,
      "max_input_tokens": 272000,
      "available_providers": ["bedrock"]
    },
    "openai.gpt-5.4": {
      "format": "openai",
      "flavor": "chat",
      "multimodal": true,
      "input_cost_per_mil_tokens": 2.75,
      "output_cost_per_mil_tokens": 16.5,
      "input_cache_read_cost_per_mil_tokens": 0.275,
      "displayName": "OpenAI GPT-5.4 (Bedrock)",
      "reasoning": true,
      "max_input_tokens": 272000,
      "available_providers": ["bedrock"]
    }
  },
  "source_urls": [
    "https://docs.aws.amazon.com/bedrock/latest/userguide/models-supported.html",
    "https://aws.amazon.com/bedrock/pricing/"
  ]
}

[github-actions]

Prepared catalog updates for openai.gpt-5.5, openai.gpt-5.4. Added models: openai.gpt-5.5, openai.gpt-5.4.

Verification sources: https://docs.aws.amazon.com/bedrock/latest/userguide/models-supported.html, https://aws.amazon.com/bedrock/pricing/, https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-openai-gpt-55.html, https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-openai-gpt-54.html
Verified metadata applied:

• openai.gpt-5.5: format=openai, flavor=chat, displayName=OpenAI GPT-5.5 (Bedrock), max_input_tokens=272000, max_output_tokens=not provided, pricing(input/output per 1M)=5.5/33, input_cache_read_cost_per_mil_tokens=0.55, providers=bedrock, reasoning=true, multimodal=true
• openai.gpt-5.4: format=openai, flavor=chat, displayName=OpenAI GPT-5.4 (Bedrock), max_input_tokens=272000, max_output_tokens=not provided, pricing(input/output per 1M)=2.75/16.5, input_cache_read_cost_per_mil_tokens=0.275, providers=bedrock, reasoning=true, multimodal=true

Included in the daily batch PR: #822