Gemini 2.5 Flash Preview 09-2025 is the September preview edition of Google's hybrid reasoning model. Its documented inputs are text, images, video and audio; the output is text. The model can combine the question with visual or spoken material, then produce an explanation, a document, code or a structured response. This exact dated edition is distinct from stable Gemini 2.5 Flash and from a moving latest alias. Its purpose is multimodal understanding and tool-assisted work with a written result.
Google describes two specific improvements for this release: better agentic tool use and more economical answer generation when thinking is enabled. The first concerns tasks that need several connected actions, such as obtaining information, comparing it with the request and choosing the next step. The second concerns producing useful answers with fewer generated tokens while retaining quality. For a development assistant, supply the observed behaviour, relevant code and the criterion for a successful change. For a research assistant, define the question, the material to inspect and the evidence needed for the conclusion.
Thinking and direct response are part of the hybrid design. A thinking task can analyse dependencies, mathematical conditions or a multi-step plan before preparing the final answer. A straightforward extraction or short explanation can have a simpler response objective. State which constraints are mandatory, what the model may assume and what the final artifact must contain. The manufacturer's family card notes that adherence to thinking budgets can be inconsistent, so a requested budget is not proof of a strict reasoning ceiling. This description does not attach a newer model's thinking levels or defaults to the September edition.
The exact September model table specifies up to 1,048,576 input tokens and 65,536 output tokens. Its large input profile can hold instructions, examples, reference text and multimedia context in a single task. To compare requirements with a proposed solution, label the two sources and provide a common set of criteria. To discuss a screenshot, identify the relevant state or component; for an audio or video task, name the question or event rather than requesting an undirected description. The result remains text: native image creation, voice generation and real-time spoken dialogue belong to other model profiles.
Google documents function calling, code execution, search grounding, URL context and file search for this exact preview. Function calling chooses a declared function and prepares arguments; the connected application performs the operation and returns its result. Code execution uses generated Python for calculations or text processing. Search grounding supplies web material for a source-informed answer, while URL context works with the contents of specified accessible pages. File search provides another source-retrieval capability. Give an agent a clear purpose for each tool and specify the expected information from every step, so the workflow can continue from the actual returned result.
Structured output support makes the final artifact usable as a set of defined fields. For a comparison, request a criterion, the corresponding source passage, the observed difference and the question that still needs clarification. For a code task, separate the proposed change from its explanation and the checks still required. Google's exact model profile also lists context caching and batch processing support. Repeated source material can be kept consistent across related tasks; caching does not itself establish that any particular request reused cached tokens. No newer schema combinations, cache thresholds or simultaneous tool modes are implied by these capability names.
The family model card describes a sparse mixture-of-experts transformer with native text, visual and audio understanding. Training combines web text, code, images, audio and video; post-training includes paired multimodal instructions and responses, human preferences and tool-use data. Data preparation applies deduplication, safety and quality filtering. The exact dated model page gives January 2025 as its knowledge cutoff. Questions requiring newer information should identify their source material or an appropriate retrieval step rather than treat the model's training knowledge as a live information feed.
Substantive limitations include possible factual invention and difficulties with causal understanding, complex logical deduction and counterfactual reasoning. The family card also notes that responses can sometimes have an unnecessarily lecturing tone. For a useful result, require the model to distinguish source facts, assumptions and proposed actions, then inspect the portions that determine the decision. A complete agentic artifact might contain the findings, a justified next step, the information obtained from tools and any unresolved question. The September release's tool-use emphasis helps define that workflow, while the prepared inputs and explicit completion criteria keep it tied to the user's task.
", "aboutText": "Gemini 2.5 Flash Preview 09-2025 is the September preview edition of Google's hybrid reasoning model. Its documented inputs are text, images, video and audio; the output is text. The model can combine the question with visual or spoken material, then produce an explanation, a document, code or a structured response. This exact dated edition is distinct from stable Gemini 2.5 Flash and from a moving latest alias. Its purpose is multimodal understanding and tool-assisted work with a written result.\n\nGoogle describes two specific improvements for this release: better agentic tool use and more economical answer generation when thinking is enabled. The first concerns tasks that need several connected actions, such as obtaining information, comparing it with the request and choosing the next step. The second concerns producing useful answers with fewer generated tokens while retaining quality. For a development assistant, supply the observed behaviour, relevant code and the criterion for a successful change. For a research assistant, define the question, the material to inspect and the evidence needed for the conclusion.\n\nThinking and direct response are part of the hybrid design. A thinking task can analyse dependencies, mathematical conditions or a multi-step plan before preparing the final answer. A straightforward extraction or short explanation can have a simpler response objective. State which constraints are mandatory, what the model may assume and what the final artifact must contain. The manufacturer's family card notes that adherence to thinking budgets can be inconsistent, so a requested budget is not proof of a strict reasoning ceiling. This description does not attach a newer model's thinking levels or defaults to the September edition.\n\nThe exact September model table specifies up to 1,048,576 input tokens and 65,536 output tokens. Its large input profile can hold instructions, examples, reference text and multimedia context in a single task. To compare requirements with a proposed solution, label the two sources and provide a common set of criteria. To discuss a screenshot, identify the relevant state or component; for an audio or video task, name the question or event rather than requesting an undirected description. The result remains text: native image creation, voice generation and real-time spoken dialogue belong to other model profiles.\n\nGoogle documents function calling, code execution, search grounding, URL context and file search for this exact preview. Function calling chooses a declared function and prepares arguments; the connected application performs the operation and returns its result. Code execution uses generated Python for calculations or text processing. Search grounding supplies web material for a source-informed answer, while URL context works with the contents of specified accessible pages. File search provides another source-retrieval capability. Give an agent a clear purpose for each tool and specify the expected information from every step, so the workflow can continue from the actual returned result.\n\nStructured output support makes the final artifact usable as a set of defined fields. For a comparison, request a criterion, the corresponding source passage, the observed difference and the question that still needs clarification. For a code task, separate the proposed change from its explanation and the checks still required. Google's exact model profile also lists context caching and batch processing support. Repeated source material can be kept consistent across related tasks; caching does not itself establish that any particular request reused cached tokens. No newer schema combinations, cache thresholds or simultaneous tool modes are implied by these capability names.\n\nThe family model card describes a sparse mixture-of-experts transformer with native text, visual and audio understanding. Training combines web text, code, images, audio and video; post-training includes paired multimodal instructions and responses, human preferences and tool-use data. Data preparation applies deduplication, safety and quality filtering. The exact dated model page gives January 2025 as its knowledge cutoff. Questions requiring newer information should identify their source material or an appropriate retrieval step rather than treat the model's training knowledge as a live information feed.\n\nSubstantive limitations include possible factual invention and difficulties with causal understanding, complex logical deduction and counterfactual reasoning. The family card also notes that responses can sometimes have an unnecessarily lecturing tone. For a useful result, require the model to distinguish source facts, assumptions and proposed actions, then inspect the portions that determine the decision. A complete agentic artifact might contain the findings, a justified next step, the information obtained from tools and any unresolved question. The September release's tool-use emphasis helps define that workflow, while the prepared inputs and explicit completion criteria keep it tied to the user's task.", "editorialOverlay": { "copy_fields": { "practical_html": "Technical example in English
Review a proposed retention design\n\nAssess this design against the requirement. Requirement R: delete inactive guest records after 30 days; paid accounts are excluded. Design D: a nightly job deletes every account whose last_login is older than 30 days. List mismatches with R/D references, propose a corrected selection rule in plain English, and give boundary cases to test.A concise requirements review, a corrected guest-only rule and tests for day 30, day 31 and an inactive paid account.The proposal excludes paid accounts and explicitly resolves what after 30 days means at the boundary.gemini-2.5-flash-preview-09-2025Gemini 2.5 Flash Preview 09-2025 is Google’s September edition for multimodal analysis and multi-step work with tools. It combines text, image, video and audio understanding with thinking, structured responses and a 1,048,576-token input limit, with specific improvements in agentic tool use.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Gemini 2.5 Flash Preview 09-2025 is the September preview edition of Google's hybrid reasoning model. Its documented inputs are text, images, video and audio; the output is text. The model can combine the question with visual or spoken material, then produce an explanation, a document, code or a structured response. This exact dated edition is distinct from stable Gemini 2.5 Flash and from a moving latest alias. Its purpose is multimodal understanding and tool-assisted work with a written result.
Google describes two specific improvements for this release: better agentic tool use and more economical answer generation when thinking is enabled. The first concerns tasks that need several connected actions, such as obtaining information, comparing it with the request and choosing the next step. The second concerns producing useful answers with fewer generated tokens while retaining quality. For a development assistant, supply the observed behaviour, relevant code and the criterion for a successful change. For a research assistant, define the question, the material to inspect and the evidence needed for the conclusion.
Thinking and direct response are part of the hybrid design. A thinking task can analyse dependencies, mathematical conditions or a multi-step plan before preparing the final answer. A straightforward extraction or short explanation can have a simpler response objective. State which constraints are mandatory, what the model may assume and what the final artifact must contain. The manufacturer's family card notes that adherence to thinking budgets can be inconsistent, so a requested budget is not proof of a strict reasoning ceiling. This description does not attach a newer model's thinking levels or defaults to the September edition.
The exact September model table specifies up to 1,048,576 input tokens and 65,536 output tokens. Its large input profile can hold instructions, examples, reference text and multimedia context in a single task. To compare requirements with a proposed solution, label the two sources and provide a common set of criteria. To discuss a screenshot, identify the relevant state or component; for an audio or video task, name the question or event rather than requesting an undirected description. The result remains text: native image creation, voice generation and real-time spoken dialogue belong to other model profiles.
Google documents function calling, code execution, search grounding, URL context and file search for this exact preview. Function calling chooses a declared function and prepares arguments; the connected application performs the operation and returns its result. Code execution uses generated Python for calculations or text processing. Search grounding supplies web material for a source-informed answer, while URL context works with the contents of specified accessible pages. File search provides another source-retrieval capability. Give an agent a clear purpose for each tool and specify the expected information from every step, so the workflow can continue from the actual returned result.
Structured output support makes the final artifact usable as a set of defined fields. For a comparison, request a criterion, the corresponding source passage, the observed difference and the question that still needs clarification. For a code task, separate the proposed change from its explanation and the checks still required. Google's exact model profile also lists context caching and batch processing support. Repeated source material can be kept consistent across related tasks; caching does not itself establish that any particular request reused cached tokens. No newer schema combinations, cache thresholds or simultaneous tool modes are implied by these capability names.
The family model card describes a sparse mixture-of-experts transformer with native text, visual and audio understanding. Training combines web text, code, images, audio and video; post-training includes paired multimodal instructions and responses, human preferences and tool-use data. Data preparation applies deduplication, safety and quality filtering. The exact dated model page gives January 2025 as its knowledge cutoff. Questions requiring newer information should identify their source material or an appropriate retrieval step rather than treat the model's training knowledge as a live information feed.
Substantive limitations include possible factual invention and difficulties with causal understanding, complex logical deduction and counterfactual reasoning. The family card also notes that responses can sometimes have an unnecessarily lecturing tone. For a useful result, require the model to distinguish source facts, assumptions and proposed actions, then inspect the portions that determine the decision. A complete agentic artifact might contain the findings, a justified next step, the information obtained from tools and any unresolved question. The September release's tool-use emphasis helps define that workflow, while the prepared inputs and explicit completion criteria keep it tied to the user's task.
POST /v1/chat/completionstexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits gemini-2.5-flash-preview-09-2025 | Default— |
Fieldmessages | Typemessage[] | RequirementRequired | Values & limits Minimum items: 1 | Default— |
Fieldmessages[].role | Typestring | RequirementRequired | Values & limits Examples: system, user, assistant, tool, function Conditions: These roles are explicitly represented by Source compatibility types; the exact model role set remains model-specific. Model-specific | Default— |
Fieldmessages[].content | Typestring|content_part[]|null | RequirementOptional | Values & limits | Default— |
Fieldmessages[].tool_calls | Typetool_call[] | RequirementOptional | Values & limits | Default— |
Fieldmessages[].tool_call_id | Typestring | RequirementOptional | Values & limits Applies when: role=tool; use actual returned ID | Default— |
Fieldmessages[].name | Typestring | RequirementOptional | Values & limits | Default— |
Fieldmessages[].reasoning_content | Typestring | RequirementOptional | Values & limits Applies when: Round-trip actual returned reasoning where this model uses it | Default— |
Fieldmessages[].reasoning_details | Typeobject|array | RequirementOptional | Values & limits Model-specific | Default— |
Fieldmax_tokens | Typeinteger | RequirementOptional | Values & limits Minimum: 1 Model-specific | Default— |
Fieldmax_completion_tokens | Typeinteger | RequirementOptional | Values & limits Minimum: 1 Model-specific | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits | Defaultfalse |
Fieldstream_options.include_usage | Typeboolean | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtemperature | Typenumber | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtop_p | Typenumber | RequirementOptional | Values & limits Model-specific | Default— |
Fieldstop | Typestring|string[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtools | Typefunction_tool[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtools[].type | Typestring | RequirementOptional | Values & limits function Model-specific | Default— |
Fieldtools[].function.name | Typestring | RequirementRequired | Values & limits Applies when: function tool supplied | Default— |
Fieldtools[].function.description | Typestring | RequirementOptional | Values & limits | Default— |
Fieldtools[].function.parameters | TypeJSON_schema | RequirementOptional | Values & limits | Default— |
Fieldtool_choice | Typestring|object | RequirementOptional | Values & limits Model-specific | Default— |
Fieldparallel_tool_calls | Typeboolean | RequirementOptional | Values & limits Model-specific | Default— |
Fieldresponse_format | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Fieldreasoning_effort | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldreasoning | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Fieldservice_tier | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldinstructions | Typestring | RequirementOptional | Values & limits Compatibility field | Default— |
Fieldfunctions | Typefunction[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldfunction_call | Typestring|object | RequirementOptional | Values & limits Model-specific | Default— |
Fieldprompt_cache_options | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Fieldmodalities | Typestring[] | RequirementOptional | Values & limits text | Default— |
Fieldprovider | Typeobject | RequirementOptional | Values & limits Conditions: provider.options is rejected on this Ultra text route | Default— |
Fieldcache_control.ttl | Typestring | RequirementOptional | Values & limits Conditions: 1h is rejected in actual prompt-cache locations | Default— |
Text · UTF8 text
messages[].content
| Parameter | Contract |
|---|---|
Parametercontent_type | Contractapplication/json |
Parameterrequired | Contractmodel, messages |
Parametermaterial_types | Contracttext |
Parametermessages | Contractmessage array with role/content shape |
Parameterfile_material | Contractactual UTF-8 document inserted as literal text; no file ID inferred |
Parameteroptional_controls | Contractstream, temperature, top_p, stop, max_tokens, max_completion_tokens, reasoning, response_format, tools, tool_choice |
Parametercontrol_scope | Contractmodel-specific; user function tools differ from hosted tools |
Parameteroutput_modalities | Contracttext |
Parameterhosted_operations | Contractfalse |
Parametermultimodal_input_binding | Contractnull |
Lifecycle: synchronous_or_streamed
Send exact model ID with messages.
Context length is not an output cap.
A text file may supply literal text; image/audio/file-object inputs need exact binding.
Tariff references do not add request options. Supported parameters are listed for the selected method.
| Configuration | Rate schedule | Charge |
|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Charge Input · Input tokens $0.0006 / 1K tokens Output · Output tokens $0.005 / 1K tokens Input · Cache read $0 / 1K tokens Input · Cache write $0 / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np=json.loads('{"model": "gemini-2.5-flash-preview-09-2025", "messages": [{"role": "user", "content": "Explain the difference between an observation and a hypothesis in three sentences."}]}')\n\nPath("request.json").write_text(json.dumps(p,ensure_ascii=False))\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/chat/completions \\\n -H "Authorization: Bearer $AILOOT_API_KEY" -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o result.json\npython3 - <<'PY_RESULT'\nimport json\nfrom pathlib import Path\nr=json.loads(Path("result.json").read_text())\nfor x in r.get("choices",[]):\n if x.get("message",{}).get("content"): print(x["message"]["content"])\nPY_RESULT\nfrom openai import OpenAI\nimport os\n\nif not os.environ.get("AILOOT_API_KEY", "").strip(): raise SystemExit("Set AILOOT_API_KEY before making a request")\nclient=OpenAI(base_url="https://api.ailoot.ai/v1",api_key=os.environ["AILOOT_API_KEY"],max_retries=0)\nresult=client.chat.completions.create(model='gemini-2.5-flash-preview-09-2025',messages=[{"role":"user","content":"Explain observations and hypotheses in three sentences."}])\nimport base64, json\nfrom pathlib import Path\n\ndef save_result(data, stem="result"):\n # Keep every field, including usage, IDs, reasoning and tool requests.\n Path(stem + ".json").write_text(json.dumps(data, ensure_ascii=False, indent=2), encoding="utf-8")\n print(json.dumps(data, ensure_ascii=False, indent=2))\n def save_blob(encoded, suffix, label):\n raw = base64.b64decode(encoded, altchars=b"-_", validate=True)\n ext = "png" if raw.startswith(b"\\x89PNG\\r\\n\\x1a\\n") else "jpg" if raw.startswith(b"\\xff\\xd8") else "webp" if raw[:4] == b"RIFF" and raw[8:12] == b"WEBP" else suffix\n file = Path(stem + "-" + label + "." + ext)\n file.write_bytes(raw)\n print("Saved", file)\n for index, block in enumerate(data.get("content") or []):\n print("Content", index, json.dumps(block, ensure_ascii=False))\n for choice in data.get("choices") or []:\n print("Choice", choice.get("index"), json.dumps(choice.get("message", choice.get("delta", {})), ensure_ascii=False))\n for index, item in enumerate(data.get("output") or []):\n print("Output", index, json.dumps(item, ensure_ascii=False))\n if item.get("type") == "image_generation_call" and item.get("result"):\n save_blob(item["result"], item.get("output_format") or "bin", "output-" + str(index))\n for ci, candidate in enumerate(data.get("candidates") or []):\n for pi, part in enumerate((candidate.get("content") or {}).get("parts") or []):\n print("Candidate/part", ci, pi, json.dumps(part, ensure_ascii=False))\n blob = part.get("inlineData") or part.get("inline_data")\n if blob and blob.get("data"):\n mime = blob.get("mimeType") or blob.get("mime_type")\n suffix = {"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"}.get(mime, "bin")\n save_blob(blob["data"], suffix, "candidate-" + str(ci) + "-part-" + str(pi))\n\nsave_result(result.to_dict(mode="json"))\nfrom openai import OpenAI\nimport os\nfrom pathlib import Path\n\nif not os.environ.get("AILOOT_API_KEY", "").strip(): raise SystemExit("Set AILOOT_API_KEY before making a request")\nclient=OpenAI(base_url="https://api.ailoot.ai/v1",api_key=os.environ["AILOOT_API_KEY"],max_retries=0)\nimport base64, json\nfrom pathlib import Path\n\ndef save_result(data, stem="result"):\n # Keep every field, including usage, IDs, reasoning and tool requests.\n Path(stem + ".json").write_text(json.dumps(data, ensure_ascii=False, indent=2), encoding="utf-8")\n print(json.dumps(data, ensure_ascii=False, indent=2))\n def save_blob(encoded, suffix, label):\n raw = base64.b64decode(encoded, altchars=b"-_", validate=True)\n ext = "png" if raw.startswith(b"\\x89PNG\\r\\n\\x1a\\n") else "jpg" if raw.startswith(b"\\xff\\xd8") else "webp" if raw[:4] == b"RIFF" and raw[8:12] == b"WEBP" else suffix\n file = Path(stem + "-" + label + "." + ext)\n file.write_bytes(raw)\n print("Saved", file)\n for index, block in enumerate(data.get("content") or []):\n print("Content", index, json.dumps(block, ensure_ascii=False))\n for choice in data.get("choices") or []:\n print("Choice", choice.get("index"), json.dumps(choice.get("message", choice.get("delta", {})), ensure_ascii=False))\n for index, item in enumerate(data.get("output") or []):\n print("Output", index, json.dumps(item, ensure_ascii=False))\n if item.get("type") == "image_generation_call" and item.get("result"):\n save_blob(item["result"], item.get("output_format") or "bin", "output-" + str(index))\n for ci, candidate in enumerate(data.get("candidates") or []):\n for pi, part in enumerate((candidate.get("content") or {}).get("parts") or []):\n print("Candidate/part", ci, pi, json.dumps(part, ensure_ascii=False))\n blob = part.get("inlineData") or part.get("inline_data")\n if blob and blob.get("data"):\n mime = blob.get("mimeType") or blob.get("mime_type")\n suffix = {"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"}.get(mime, "bin")\n save_blob(blob["data"], suffix, "candidate-" + str(ci) + "-part-" + str(pi))\n\nwith client.chat.completions.stream(model='gemini-2.5-flash-preview-09-2025',messages=[{"role":"user","content":"Explain observations and hypotheses in three sentences."}]) as stream:\n with Path("stream-events.jsonl").open("w",encoding="utf-8") as log:\n for event in stream:\n payload=event.to_dict(mode="json")\n log.write(json.dumps(payload,ensure_ascii=False)+"\\n"); log.flush()\n print(json.dumps(payload,ensure_ascii=False))\n final=stream.get_final_completion()\n save_result(final.to_dict(mode="json"),"stream-result")\njson or SSE
Keep every complete returned response, file and content part.
Keep every choice and its content/tool calls/reasoning. In SSE assemble deltas by choice/tool index, retain the final finish reason and usage.
Keep request/response IDs and headers. No generic text job retrieval URL is supplied.
| Field | Type |
|---|---|
Fieldchoices[].index | Typeinteger |
Fieldchoices[].message.content | Typestring|null |
Fieldchoices[].message.tool_calls | Typetool_call[] |
Fieldchoices[].message.reasoning_content | Typestring |
Fieldchoices[].message.reasoning_details | Typeobject|array |
Fieldchoices[].finish_reason | Typestring|null |
Fieldusage | Typeobject |
| Response | Condition | Next step |
|---|---|---|
Response400 · invalid_request_error | ConditionInvalid required fields, material representation or parameter value | Next stepCorrect the indicated fields before a new creation request. |
Ailoot Ultra gemini-2.5-flash-preview-09-2025
Use the method paths shown above.
gemini-2.5-flash-preview-09-2025Gemini 2.5 Flash Preview 09-2025 is Google’s September edition for multimodal analysis and multi-step work with tools. It combines text, image, video and audio understanding with thinking, structured responses and a 1,048,576-token input limit, with specific improvements in agentic tool use.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Gemini 2.5 Flash Preview 09-2025 is the September preview edition of Google's hybrid reasoning model. Its documented inputs are text, images, video and audio; the output is text. The model can combine the question with visual or spoken material, then produce an explanation, a document, code or a structured response. This exact dated edition is distinct from stable Gemini 2.5 Flash and from a moving latest alias. Its purpose is multimodal understanding and tool-assisted work with a written result.
Google describes two specific improvements for this release: better agentic tool use and more economical answer generation when thinking is enabled. The first concerns tasks that need several connected actions, such as obtaining information, comparing it with the request and choosing the next step. The second concerns producing useful answers with fewer generated tokens while retaining quality. For a development assistant, supply the observed behaviour, relevant code and the criterion for a successful change. For a research assistant, define the question, the material to inspect and the evidence needed for the conclusion.
Thinking and direct response are part of the hybrid design. A thinking task can analyse dependencies, mathematical conditions or a multi-step plan before preparing the final answer. A straightforward extraction or short explanation can have a simpler response objective. State which constraints are mandatory, what the model may assume and what the final artifact must contain. The manufacturer's family card notes that adherence to thinking budgets can be inconsistent, so a requested budget is not proof of a strict reasoning ceiling. This description does not attach a newer model's thinking levels or defaults to the September edition.
The exact September model table specifies up to 1,048,576 input tokens and 65,536 output tokens. Its large input profile can hold instructions, examples, reference text and multimedia context in a single task. To compare requirements with a proposed solution, label the two sources and provide a common set of criteria. To discuss a screenshot, identify the relevant state or component; for an audio or video task, name the question or event rather than requesting an undirected description. The result remains text: native image creation, voice generation and real-time spoken dialogue belong to other model profiles.
Google documents function calling, code execution, search grounding, URL context and file search for this exact preview. Function calling chooses a declared function and prepares arguments; the connected application performs the operation and returns its result. Code execution uses generated Python for calculations or text processing. Search grounding supplies web material for a source-informed answer, while URL context works with the contents of specified accessible pages. File search provides another source-retrieval capability. Give an agent a clear purpose for each tool and specify the expected information from every step, so the workflow can continue from the actual returned result.
Structured output support makes the final artifact usable as a set of defined fields. For a comparison, request a criterion, the corresponding source passage, the observed difference and the question that still needs clarification. For a code task, separate the proposed change from its explanation and the checks still required. Google's exact model profile also lists context caching and batch processing support. Repeated source material can be kept consistent across related tasks; caching does not itself establish that any particular request reused cached tokens. No newer schema combinations, cache thresholds or simultaneous tool modes are implied by these capability names.
The family model card describes a sparse mixture-of-experts transformer with native text, visual and audio understanding. Training combines web text, code, images, audio and video; post-training includes paired multimodal instructions and responses, human preferences and tool-use data. Data preparation applies deduplication, safety and quality filtering. The exact dated model page gives January 2025 as its knowledge cutoff. Questions requiring newer information should identify their source material or an appropriate retrieval step rather than treat the model's training knowledge as a live information feed.
Substantive limitations include possible factual invention and difficulties with causal understanding, complex logical deduction and counterfactual reasoning. The family card also notes that responses can sometimes have an unnecessarily lecturing tone. For a useful result, require the model to distinguish source facts, assumptions and proposed actions, then inspect the portions that determine the decision. A complete agentic artifact might contain the findings, a justified next step, the information obtained from tools and any unresolved question. The September release's tool-use emphasis helps define that workflow, while the prepared inputs and explicit completion criteria keep it tied to the user's task.
POST /v1/messagestextjson, sse| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits gemini-2.5-flash-preview-09-2025 | Default— |
Fieldmessages | Typemessage[] | RequirementRequired | Values & limits | Default— |
Fieldmessages[].role | Typestring | RequirementRequired | Values & limits user, assistant | Default— |
Fieldmessages[].content | Typestring | RequirementRequired | Values & limits | Default— |
Fieldmax_tokens | Typeinteger | RequirementRequired | Values & limits | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits | Defaultfalse |
Text · UTF-8 text
messages[].content string
| Parameter | Contract |
|---|---|
Parametercontent_type | Contractapplication/json |
Parameterrequired | Contractmodel, messages, max_tokens |
Parameteroptional_controls | Contractstream |
Parametermaterial_types | Contracttext |
Parametertext_file_material | ContractRead the actual UTF-8 document and put its text in the indicated input or message content string. |
Lifecycle: synchronous_or_streamed
Tariff references do not add request options. Supported parameters are listed for the selected method.
| Configuration | Rate schedule | Charge |
|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Charge Input · Input tokens $0.0006 / 1K tokens Output · Output tokens $0.005 / 1K tokens Input · Cache read $0 / 1K tokens Input · Cache write $0 / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
{\n "model": "gemini-2.5-flash-preview-09-2025",\n "max_tokens": 128,\n "messages": [\n {\n "role": "user",\n "content": "Summarize the incident report."\n }\n ],\n "stream": false\n}set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\n: "${AILOOT_TEXT_FILE:?Set your actual UTF-8 document}"\nexport AILOOT_TEXT_FILE\npython3 - <<'PY_REQUEST'\nimport json, os\nfrom pathlib import Path\n\nsource_file = Path(os.environ['AILOOT_TEXT_FILE'])\nif not source_file.is_file():\n raise SystemExit('Set AILOOT_TEXT_FILE to your actual UTF-8 document')\ntext = source_file.read_text(encoding='utf-8')\nif not text.strip():\n raise SystemExit('Choose a document containing text')\npayload = {'model': 'gemini-2.5-flash-preview-09-2025', 'max_tokens': 128, 'messages': [{'role': 'user', 'content': 'Summarize the incident report.'}]}\npayload['messages'][0]['content'] = 'Summarize this material and quote supporting source passages.\\n\\n' + text\nPath('request.json').write_text(json.dumps(payload, ensure_ascii=False), encoding='utf-8')\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/messages \\\n -H "x-api-key: $AILOOT_API_KEY" -H 'anthropic-version: 2023-06-01' -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o result.json\npython3 - <<'PY_RESULT'\nimport json\nfrom pathlib import Path\n\nresult = json.loads(Path('result.json').read_text(encoding='utf-8'))\nprint(json.dumps(result, ensure_ascii=False, indent=2))\nPY_RESULT\nimport json, os\nfrom pathlib import Path\n\nsource_file = Path(os.environ['AILOOT_TEXT_FILE'])\nif not source_file.is_file():\n raise SystemExit('Set AILOOT_TEXT_FILE to your actual UTF-8 document')\ntext = source_file.read_text(encoding='utf-8')\nif not text.strip():\n raise SystemExit('Choose a document containing text')\npayload = {'model': 'gemini-2.5-flash-preview-09-2025', 'max_tokens': 128, 'messages': [{'role': 'user', 'content': 'Summarize the incident report.'}]}\npayload['messages'][0]['content'] = 'Summarize this material and quote supporting source passages.\\n\\n' + text\nfrom urllib.request import Request, urlopen\nfrom urllib.error import HTTPError\n\napi_key = os.environ['AILOOT_API_KEY']\nif not api_key.strip():\n raise SystemExit('Set AILOOT_API_KEY before making a request')\nheaders = {'x-api-key': 'PLACEHOLDER_KEY', 'anthropic-version': '2023-06-01', 'Content-Type': 'application/json'}\nheaders['x-api-key'] = api_key\nrequest = Request('https://api.ailoot.ai/v1/messages', data=json.dumps(payload, ensure_ascii=False).encode('utf-8'), headers=headers, method='POST')\ntry:\n with urlopen(request, timeout=120) as response:\n raw = response.read()\n Path('response-headers.txt').write_text(str(response.headers), encoding='utf-8')\nexcept HTTPError as error:\n Path('result.json').write_bytes(error.read())\n Path('response-headers.txt').write_text(str(error.headers), encoding='utf-8')\n raise\nPath('result.json').write_bytes(raw)\nimport json\nfrom pathlib import Path\n\nresult = json.loads(Path('result.json').read_text(encoding='utf-8'))\nprint(json.dumps(result, ensure_ascii=False, indent=2))\nJSON or SSE
Keep every complete returned response, file and content part.
Keep all content blocks, thinking signatures and actual tool-use IDs. Assemble all SSE block deltas; a first text block is not the complete output.
Keep request/response IDs and headers. No generic text job retrieval URL is supplied.
| Field | Type |
|---|---|
Fieldid | Typestring |
Fieldmodel | Typestring |
Fieldcontent[] | Typetext|thinking|tool_use|other content block |
Fieldstop_reason | Typestring|null |
Fieldusage | Typeobject |
| Response | Condition | Next step |
|---|---|---|
Response400 · invalid_request_error | ConditionInvalid required fields, material representation or parameter value | Next stepCorrect the indicated fields before a new creation request. |
Ailoot Ultra gemini-2.5-flash-preview-09-2025
Use the method paths shown above.
Technical example in English
Review a proposed retention design\n\nAssess this design against the requirement. Requirement R: delete inactive guest records after 30 days; paid accounts are excluded. Design D: a nightly job deletes every account whose last_login is older than 30 days. List mismatches with R/D references, propose a corrected selection rule in plain English, and give boundary cases to test.A concise requirements review, a corrected guest-only rule and tests for day 30, day 31 and an inactive paid account.The proposal excludes paid accounts and explicitly resolves what after 30 days means at the boundary.