{
  "schema_version": "1.0",
  "publisher": "AI Release",
  "article": {
    "title": "Responses API and gpt-image-2: reasoning.effort runs into complexity",
    "canonical_url": "https://ai-release.org/posts/responses-api-i-gpt-image-2-reasoning-effort-upiraetsya-v-complexity-2e644645_en.html",
    "language": "en",
    "article_type": "analysis",
    "published_at": "2026-11-19T00:00:00+00:00",
    "updated_at": "2026-11-19T00:00:00+00:00"
  },
  "summary": "An OpenAI Community thread is dissecting a Responses API breakage: after gpt-image-2.6 shipped, requests to the image_generation tool with an explicit reasoning.effort started failing with HTTP 400 and the message \"Complexity 'low' is not supported for this image model.\" The thread's author, henry_m, reproduced it on gpt-6-luna with the gpt-image-2 tool and posted the full script. What got under everyone's skin is that they changed nothing on their end: nobody sends a complexity parameter in the request — the API injects it itself.",
  "topics": [
    "AI",
    "OpenAI",
    "API"
  ],
  "entities": [
    {
      "id": "E-SRC",
      "name": "community.openai.com",
      "type": "source",
      "description": "source domain",
      "url": "https://community.openai.com/t/responses-api-with-gpt-image-2-gives-complexity-low-is-not-supported-after-gpt-image-2-6-launch-with-temporary-hack/1412952"
    },
    {
      "id": "E-T1",
      "name": "AI",
      "type": "topic",
      "description": "publication tag",
      "url": null
    },
    {
      "id": "E-T2",
      "name": "OpenAI",
      "type": "topic",
      "description": "publication tag",
      "url": null
    },
    {
      "id": "E-T3",
      "name": "API",
      "type": "topic",
      "description": "publication tag",
      "url": null
    },
    {
      "id": "E-AIR",
      "name": "AI Release",
      "type": "publisher",
      "description": "published the article",
      "url": "https://ai-release.org/"
    }
  ],
  "claims": [
    {
      "id": "AR-RESPONSESA-001",
      "statement": "An OpenAI Community thread is dissecting a Responses API breakage: after gpt-image-2.6 shipped, requests to the image_generation tool with an explicit reasoning.effort started failing with HTTP 400 and the message \"Complexity 'low' is not supported for this image model.\" The thread's author, henry_m, reproduced it on gpt-6-luna with the gpt-image-2 tool and posted the full script.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://community.openai.com/t/responses-api-with-gpt-image-2-gives-complexity-low-is-not-supported-after-gpt-image-2-6-launch-with-temporary-hack/1412952",
      "article_anchor": "#ai-claim-001",
      "evidence_excerpt": "An OpenAI Community thread is dissecting a Responses API breakage: after gpt-image-2.6 shipped, requests to the image_generation tool with an explicit reasoning.effort started failing with HTTP 400 and the message \"Complexity 'low' is not supported for this image model.\" The thread's author, henry_m, reproduced it on gpt-6-luna with the gpt-image-2 tool and posted the full script."
    },
    {
      "id": "AR-RESPONSESA-002",
      "statement": "What got under everyone's skin is that they changed nothing on their end: nobody sends a complexity parameter in the request — the API injects it itself.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://community.openai.com/t/responses-api-with-gpt-image-2-gives-complexity-low-is-not-supported-after-gpt-image-2-6-launch-with-temporary-hack/1412952",
      "article_anchor": "#ai-claim-002",
      "evidence_excerpt": "What got under everyone's skin is that they changed nothing on their end: nobody sends a complexity parameter in the request — the API injects it itself."
    },
    {
      "id": "AR-RESPONSESA-003",
      "statement": "henry_m, who opened the thread, boils the symptom down to one sentence: \"the issue is that the reasoning effort is passed as 'complexity' at OpenAI's end without validating the model.\" In other words, the client asks for a low reasoning level, and the server forwards that value as image-generation complexity without checking which model is actually being called.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://community.openai.com/t/responses-api-with-gpt-image-2-gives-complexity-low-is-not-supported-after-gpt-image-2-6-launch-with-temporary-hack/1412952",
      "article_anchor": "#ai-claim-003",
      "evidence_excerpt": "henry_m, who opened the thread, boils the symptom down to one sentence: \"the issue is that the reasoning effort is passed as 'complexity' at OpenAI's end without validating the model.\" In other words, the client asks for a low reasoning level, and the server forwards that value as image-generation complexity without checking which model is actually being called."
    },
    {
      "id": "AR-RESPONSESA-004",
      "statement": "A participant going by _j probed the adjacent surfaces and mapped the boundaries: \"there is no issue when not sending complexity to gpt-image-2 on the generation or edit endpoint.\" According to him, complexity is a parameter only for gpt-image-2.6, and the weirdness is in the docs themselves: they claim a default of \"high.\" If the Responses API injects \"low\" on 2.6, expectations break.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://community.openai.com/t/responses-api-with-gpt-image-2-gives-complexity-low-is-not-supported-after-gpt-image-2-6-launch-with-temporary-hack/1412952",
      "article_anchor": "#ai-claim-004",
      "evidence_excerpt": "A participant going by _j probed the adjacent surfaces and mapped the boundaries: \"there is no issue when not sending complexity to gpt-image-2 on the generation or edit endpoint.\" According to him, complexity is a parameter only for gpt-image-2.6, and the weirdness is in the docs themselves: they claim a default of \"high.\" If the Responses API injects \"low\" on 2.6, expectations break."
    },
    {
      "id": "AR-RESPONSESA-005",
      "statement": "He also notes that \"auto\" doesn't exist for the parameter — only null.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://community.openai.com/t/responses-api-with-gpt-image-2-gives-complexity-low-is-not-supported-after-gpt-image-2-6-launch-with-temporary-hack/1412952",
      "article_anchor": "#ai-claim-005",
      "evidence_excerpt": "He also notes that \"auto\" doesn't exist for the parameter — only null."
    },
    {
      "id": "AR-RESPONSESA-006",
      "statement": "sps confirms: \"I can confirm that I'm seeing the same behavior on my end\" — and says he passed the signal to the team along with thanks for the detailed repro.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://community.openai.com/t/responses-api-with-gpt-image-2-gives-complexity-low-is-not-supported-after-gpt-image-2-6-launch-with-temporary-hack/1412952",
      "article_anchor": "#ai-claim-006",
      "evidence_excerpt": "sps confirms: \"I can confirm that I'm seeing the same behavior on my end\" — and says he passed the signal to the team along with thanks for the detailed repro."
    },
    {
      "id": "AR-RESPONSESA-007",
      "statement": "It's the only message in the thread that mentions an internal escalation rather than outside observations.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://community.openai.com/t/responses-api-with-gpt-image-2-gives-complexity-low-is-not-supported-after-gpt-image-2-6-launch-with-temporary-hack/1412952",
      "article_anchor": "#ai-claim-007",
      "evidence_excerpt": "It's the only message in the thread that mentions an internal escalation rather than outside observations."
    },
    {
      "id": "AR-RESPONSESA-008",
      "statement": "rus1243mor describes his own flavor of the same problem: since GPT Image 2.6 launched (his message dates it November 4), the image_generation tool fails on gpt-image-2.5-sunburst whenever reasoning.effort is present in the request.",
      "claim_type": "reported_fact",
      "confidence": "not_assessed",
      "source_type": "publication source",
      "verification_status": "reported_unverified",
      "source_url": "https://community.openai.com/t/responses-api-with-gpt-image-2-gives-complexity-low-is-not-supported-after-gpt-image-2-6-launch-with-temporary-hack/1412952",
      "article_anchor": "#ai-claim-008",
      "evidence_excerpt": "rus1243mor describes his own flavor of the same problem: since GPT Image 2.6 launched (his message dates it November 4), the image_generation tool fails on gpt-image-2.5-sunburst whenever reasoning.effort is present in the request."
    }
  ],
  "relationships": [
    {
      "subject": "community.openai.com",
      "predicate": "reports on",
      "object": "Responses API and gpt-image-2: reasoning.effort runs into complexity",
      "claim_id": "AR-RESPONSESA-001"
    },
    {
      "subject": "AI Release",
      "predicate": "publishes",
      "object": "Responses API and gpt-image-2: reasoning.effort runs into complexity",
      "claim_id": "AR-RESPONSESA-001"
    }
  ],
  "questions_and_answers": [
    {
      "question": "What is this article about?",
      "answer": "Responses API and gpt-image-2: reasoning.effort runs into complexity. An OpenAI Community thread is dissecting a Responses API breakage: after gpt-image-2.6 shipped, requests to the image_generation tool with an explicit reasoning.effort started failing with HTTP 400 and the message \"Complexity 'low' is not supported for this image model.\" The thread's author, henry_m, reproduced it on gpt-6-luna with the gpt-image-2 tool and posted the full script.",
      "claim_ids": [
        "AR-RESPONSESA-001"
      ],
      "article_anchor": "#ai-knowledge"
    },
    {
      "question": "What happened?",
      "answer": "An OpenAI Community thread is dissecting a Responses API breakage: after gpt-image-2.6 shipped, requests to the image_generation tool with an explicit reasoning.effort started failing with HTTP 400 and the message \"Complexity 'low' is not supported for this image model.\" The thread's author, henry_m, reproduced it on gpt-6-luna with the gpt-image-2 tool and posted the full script. What got under everyone's skin is that they changed nothing on their end: nobody sends a complexity parameter in the request — the API injects it itself. henry_m, who opened the thread, boils the symptom down to one sentence: \"the issue is that the reasoning effort is passed as 'complexity' at OpenAI's end without validating the model.\" In other words, the client asks for a low reasoning level, and the server forwards that value as image-generation complexity without checking which model is actually being called.",
      "claim_ids": [
        "AR-RESPONSESA-001",
        "AR-RESPONSESA-002",
        "AR-RESPONSESA-003"
      ],
      "article_anchor": "#ai-knowledge"
    },
    {
      "question": "Who is the source?",
      "answer": "community.openai.com: https://community.openai.com/t/responses-api-with-gpt-image-2-gives-complexity-low-is-not-supported-after-gpt-image-2-6-launch-with-temporary-hack/1412952",
      "claim_ids": [],
      "article_anchor": "#ai-knowledge"
    },
    {
      "question": "What remains unknown?",
      "answer": "The article reports the cited source; independent verification and full details are not provided.",
      "claim_ids": [],
      "article_anchor": "#ai-knowledge"
    }
  ],
  "limitations": [
    "No independent verification is provided.",
    "Full details, dates and official documents may be missing.",
    "The article reports a source rather than being the primary source."
  ],
  "citation": {
    "preferred_citation": "AI Release. «Responses API and gpt-image-2: reasoning.effort runs into complexity». 2026-11-19. https://ai-release.org/posts/responses-api-i-gpt-image-2-reasoning-effort-upiraetsya-v-complexity-2e644645_en.html",
    "canonical_url": "https://ai-release.org/posts/responses-api-i-gpt-image-2-reasoning-effort-upiraetsya-v-complexity-2e644645_en.html",
    "publisher": "AI Release"
  }
}