Documentation 2026-07-11 07:01 UTC to 2026-07-13 21:58 UTC

38- This guide's examples use `us-east-2`, which resolves to38- This guide's examples use `us-east-2`, which resolves to

39 `https://bedrock-mantle.us-east-2.api.aws/openai/v1`.39 `https://bedrock-mantle.us-east-2.api.aws/openai/v1`.

40- Use a Bedrock model ID with the `openai.` prefix, such as40- Use a Bedrock model ID with the `openai.` prefix, such as

41 `openai.gpt-5.5`.41 `openai.gpt-5.6-sol`.

42 42 

43This example uses `openai.gpt-5.5` in `us-east-2`. Use a supported model and AWS43This example uses `openai.gpt-5.6-sol` in `us-east-2`. Use a supported model and

44Region combination for your Bedrock deployment.44AWS Region combination for your Bedrock deployment.

45 45 

46The following example uses a Bedrock API key stored as46The following example uses a Bedrock API key stored as

47`AWS_BEARER_TOKEN_BEDROCK`. See47`AWS_BEARER_TOKEN_BEDROCK`. See

59});59});

60 60 

61const response = await client.responses.create({61const response = await client.responses.create({

62 model: "openai.gpt-5.5",62 model: "openai.gpt-5.6-sol",

63 input: "Write a haiku about cloud infrastructure.",63 input: "Write a haiku about cloud infrastructure.",

64});64});

65 65 

72client = BedrockOpenAI(aws_region="us-east-2")72client = BedrockOpenAI(aws_region="us-east-2")

73 73 

74response = client.responses.create(74response = client.responses.create(

75 model="openai.gpt-5.5",75 model="openai.gpt-5.6-sol",

76 input="Write a haiku about cloud infrastructure.",76 input="Write a haiku about cloud infrastructure.",

77)77)

78 78 

84 -H "Content-Type: application/json" \84 -H "Content-Type: application/json" \

85 -H "Authorization: Bearer $AWS_BEARER_TOKEN_BEDROCK" \85 -H "Authorization: Bearer $AWS_BEARER_TOKEN_BEDROCK" \

86 -d '{86 -d '{

87 "model": "openai.gpt-5.5",87 "model": "openai.gpt-5.6-sol",

88 "input": "Write a haiku about cloud infrastructure."88 "input": "Write a haiku about cloud infrastructure."

89 }'89 }'

90```90```

115});115});

116 116 

117const response = await client.responses.create({117const response = await client.responses.create({

118 model: "openai.gpt-5.5",118 model: "openai.gpt-5.6-sol",

119 input: "Write a haiku about cloud infrastructure.",119 input: "Write a haiku about cloud infrastructure.",

120});120});

121 121 

132)132)

133 133 

134response = client.responses.create(134response = client.responses.create(

135 model="openai.gpt-5.5",135 model="openai.gpt-5.6-sol",

136 input="Write a haiku about cloud infrastructure.",136 input="Write a haiku about cloud infrastructure.",

137)137)

138 138 

158## Responses API feature availability158## Responses API feature availability

159 159 

160Amazon Bedrock supports a subset of Responses API capabilities available160Amazon Bedrock supports a subset of Responses API capabilities available

161through the OpenAI API. This table describes intended feature availability for161through the OpenAI API. This table describes feature availability as of the

162the initial Amazon Bedrock offering. It excludes transient availability and162date below. It excludes transient availability and service status.

163service status.

164 163 

165The information below represents feature availability as of June 8, 2026.164The information below represents feature availability as of July 13, 2026.

166 Model and Region availability can also change. For the latest information, see165 Model and Region availability can also change. For the latest information, see

167 the [AWS documentation for OpenAI models in Amazon166 the [AWS documentation for OpenAI models in Amazon

168 Bedrock](https://docs.aws.amazon.com/bedrock/latest/userguide/model-cards-openai.html)167 Bedrock](https://docs.aws.amazon.com/bedrock/latest/userguide/model-cards-openai.html)

169 and [model support by AWS168 and [model support by AWS

170 Region](https://docs.aws.amazon.com/bedrock/latest/userguide/models-region-compatibility.html).169 Region](https://docs.aws.amazon.com/bedrock/latest/userguide/models-region-compatibility.html).

171 170 

172| Capability | OpenAI API | Amazon Bedrock at initial availability |171| Capability | OpenAI API | Amazon Bedrock |

173| ------------------------- | ------------------------- | -------------------------------------- |172| ------------------------- | ----------------------------- | ------------------------------------------------- |

174| Text generation | Available | Available |173| Text generation | Available | Available |

175| Audio input | Available | Not available |174| Audio input | Available | Not available |

176| Image input | Available | Available |175| Image input | Available | Available |

179| Function calling | Available | Available |178| Function calling | Available | Available |

180| Streaming responses | Available | Available |179| Streaming responses | Available | Available |

181| WebSocket connections | Available | Not available |180| WebSocket connections | Available | Not available |

182| Context window | Model-dependent | 272,000 tokens for GPT-5.4 and GPT-5.5 |181| Context window | Model-dependent | 272,000 tokens for GPT-5.4, GPT-5.5, and GPT-5.6 |

183| Reasoning effort | Available | Available |182| Reasoning effort | Available | Available, including `max` on supported models |

184| Prompt caching | Available | Available |183| Pro mode | Available on supported models | Not available |

184| Persisted reasoning | Available on supported models | Available on supported models |

185| Prompt caching | Available | Implicit and explicit caching on supported models |

186| Programmatic Tool Calling | Available on supported models | Not available |

187| Multi-agent | Beta on supported models | Not available |

185| Custom tools | Available | Available |188| Custom tools | Available | Available |

186| Client-side `tool_search` | Available | Available |189| Client-side `tool_search` | Available | Available |

187| Hosted web search | Available | Not available |190| Hosted web search | Available | Not available |

194 197 

195Client-side `tool_search` is distinct from hosted tools and remote MCP server198Client-side `tool_search` is distinct from hosted tools and remote MCP server

196support. Hosted tools run through OpenAI-operated service infrastructure and199support. Hosted tools run through OpenAI-operated service infrastructure and

197are unavailable in the initial Amazon Bedrock offering.200are unavailable on Amazon Bedrock.

198 201 

199GPT-5.4 and GPT-5.5 have a 272,000-token context window on Amazon Bedrock.202GPT-5.4, GPT-5.5, and GPT-5.6 have a 272,000-token context window on Amazon

200Amazon Bedrock rejects requests that exceed this limit. See the AWS model cards203Bedrock. Amazon Bedrock rejects requests that exceed this limit. See the AWS

201for current model-specific limits.204model cards for current model-specific limits.

202 205 

203Treat feature parity as workload-specific. If your application depends on a206Treat feature parity as workload-specific. If your application depends on a

204specific tool, response mode, or service tier, test that behavior through207specific tool, response mode, or service tier, test that behavior through

13 13 

14Codex works best with the latest models from the GPT-5 family, such as [`gpt-5.6`](https://developers.openai.com/api/docs/models/gpt-5.6-sol). We offer a range of models specifically designed to work with coding agents like Codex, such as [`gpt-5.3-codex`](https://developers.openai.com/api/docs/models/gpt-5.3-codex), but we recommend using the latest general-purpose model for most code generation tasks.14Codex works best with the latest models from the GPT-5 family, such as [`gpt-5.6`](https://developers.openai.com/api/docs/models/gpt-5.6-sol). We offer a range of models specifically designed to work with coding agents like Codex, such as [`gpt-5.3-codex`](https://developers.openai.com/api/docs/models/gpt-5.3-codex), but we recommend using the latest general-purpose model for most code generation tasks.

15 15 

16See the [Codex docs](https://developers.openai.com/codex) for setup guides, reference material, pricing, and more information.16See the [ChatGPT docs](https://developers.openai.com/codex) for setup guides, reference material, pricing, and more information.

17 17 

18## Integrate with coding models18## Integrate with coding models

19 19 

69 69 

70## Next steps70## Next steps

71 71 

72- Visit the [Codex docs](https://developers.openai.com/codex) to learn what you can do with Codex, set up Codex in whichever interface you choose, or find more details.72- Visit the [ChatGPT docs](https://developers.openai.com/codex) to learn what you can do with Codex, set up Codex in whichever interface you choose, or find more details.

73- Read <a href="/api/docs/guides/latest-model">Model guidance</a> for model selection, features, migration guidance, and prompting patterns that work well on coding and agentic tasks.73- Read <a href="/api/docs/guides/latest-model">Model guidance</a> for model selection, features, migration guidance, and prompting patterns that work well on coding and agentic tasks.

74- Compare [`gpt-5.6`](https://developers.openai.com/api/docs/models/gpt-5.6-sol) and [`gpt-5.3-codex`](https://developers.openai.com/api/docs/models/gpt-5.3-codex) on the model pages.74- Compare [`gpt-5.6`](https://developers.openai.com/api/docs/models/gpt-5.6-sol) and [`gpt-5.3-codex`](https://developers.openai.com/api/docs/models/gpt-5.3-codex) on the model pages.