ERNIE 4.5 0.3B accepts text conversations through Ailoot Ultra. The exact model ID is baidu/ernie-4.5-0.3b. Put the source material directly in your messages and describe the result you need. This text profile suits tasks such as turning customer notes into a response, condensing a document excerpt, or extracting a specified set of facts into an agreed layout.
A useful request combines the material, the task and the answer requirements. For a support reply, supply the customer message and the policy text the response must follow. For a summary, name the intended reader and ask for decisions, outstanding questions and action items. For extraction, specify the field names and tell the model how to represent facts absent from the source. These instructions give you a result that can be checked against the material you provided.
With Chat Completions, send POST /v1/chat/completions with model and a nonempty messages array. Each message supplies a role and text content. Include the relevant previous turns when the answer depends on a conversation. To work from a document, read its actual text and place the relevant contents in the message; a filename alone does not supply the document.
With Messages, send POST /v1/messages with model, messages and max_tokens. Use user and assistant turns, with an optional separate system instruction. Content can be a text string or text blocks. max_tokens is an integer of at least 1. The published model profile lists an 8,000-token completion ceiling and a 120,000-token context; leave room for the requested answer when preparing the conversation. The context size does not mean an answer can contain 120,000 tokens.
Both interfaces provide the boolean stream control, whose default is false. For a complete JSON answer, retain every returned result field and usage record. For streamed delivery, collect the events and assemble all answer content before treating the response as complete. Chat Completions places the ordinary reply in choices[].message.content; Messages returns content blocks together with stop_reason and usage. Keep the original request identity and any response received if the connection ends before completion, so an incomplete response can be investigated without blindly sending the same request again.
", "aboutText": "ERNIE 4.5 0.3B accepts text conversations through Ailoot Ultra. The exact model ID is baidu/ernie-4.5-0.3b. Put the source material directly in your messages and describe the result you need. This text profile suits tasks such as turning customer notes into a response, condensing a document excerpt, or extracting a specified set of facts into an agreed layout.\n\nA useful request combines the material, the task and the answer requirements. For a support reply, supply the customer message and the policy text the response must follow. For a summary, name the intended reader and ask for decisions, outstanding questions and action items. For extraction, specify the field names and tell the model how to represent facts absent from the source. These instructions give you a result that can be checked against the material you provided.\n\nWith Chat Completions, send POST /v1/chat/completions with model and a nonempty messages array. Each message supplies a role and text content. Include the relevant previous turns when the answer depends on a conversation. To work from a document, read its actual text and place the relevant contents in the message; a filename alone does not supply the document.\n\nWith Messages, send POST /v1/messages with model, messages and max_tokens. Use user and assistant turns, with an optional separate system instruction. Content can be a text string or text blocks. max_tokens is an integer of at least 1. The published model profile lists an 8,000-token completion ceiling and a 120,000-token context; leave room for the requested answer when preparing the conversation. The context size does not mean an answer can contain 120,000 tokens.\n\nBoth interfaces provide the boolean stream control, whose default is false. For a complete JSON answer, retain every returned result field and usage record. For streamed delivery, collect the events and assemble all answer content before treating the response as complete. Chat Completions places the ordinary reply in choices[].message.content; Messages returns content blocks together with stop_reason and usage. Keep the original request identity and any response received if the connection ends before completion, so an incomplete response can be investigated without blindly sending the same request again.", "editorialOverlay": { "copy_fields": null, "copy_sha256": null, "copy_source": null, "html_sha256": "e1ed74efe517136bae5917df695c4fe2d6902741906ed9b1f7ec03cfa5db3770", "module_sha256": "ee9dc1c50b1e6a8fb7fa062030fcb4646b9aaaeb08be513b723a5a92a7290f06", "raw_model_unchanged": true, "schema": "ailoot-native-model-editorial-v1" }, "locale": "en", "methods": [ { "html": "baidu/ernie-4.5-0.3bERNIE 4.5 0.3B is a text-conversation model in the ERNIE 4.5 open-source family. Use the exact ID baidu/ernie-4.5-0.3b through Chat Completions or Messages to work with supplied text: draft a reply, summarize notes, extract named information, or explain a short passage. Its published profile lists a 120,000-token context and an 8,000-token completion limit; the two values describe different limits.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
ERNIE 4.5 0.3B accepts text conversations through Ailoot Ultra. The exact model ID is baidu/ernie-4.5-0.3b. Put the source material directly in your messages and describe the result you need. This text profile suits tasks such as turning customer notes into a response, condensing a document excerpt, or extracting a specified set of facts into an agreed layout.
A useful request combines the material, the task and the answer requirements. For a support reply, supply the customer message and the policy text the response must follow. For a summary, name the intended reader and ask for decisions, outstanding questions and action items. For extraction, specify the field names and tell the model how to represent facts absent from the source. These instructions give you a result that can be checked against the material you provided.
With Chat Completions, send POST /v1/chat/completions with model and a nonempty messages array. Each message supplies a role and text content. Include the relevant previous turns when the answer depends on a conversation. To work from a document, read its actual text and place the relevant contents in the message; a filename alone does not supply the document.
With Messages, send POST /v1/messages with model, messages and max_tokens. Use user and assistant turns, with an optional separate system instruction. Content can be a text string or text blocks. max_tokens is an integer of at least 1. The published model profile lists an 8,000-token completion ceiling and a 120,000-token context; leave room for the requested answer when preparing the conversation. The context size does not mean an answer can contain 120,000 tokens.
Both interfaces provide the boolean stream control, whose default is false. For a complete JSON answer, retain every returned result field and usage record. For streamed delivery, collect the events and assemble all answer content before treating the response as complete. Chat Completions places the ordinary reply in choices[].message.content; Messages returns content blocks together with stop_reason and usage. Keep the original request identity and any response received if the connection ends before completion, so an incomplete response can be investigated without blindly sending the same request again.
POST /v1/chat/completionstexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits baidu/ernie-4.5-0.3b | Default— |
Fieldmessages | Typearray | RequirementRequired | Values & limits | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits | Defaultfalse |
Fieldmax_tokens | Typeinteger | RequirementOptional | Values & limits Minimum: 1 | Default— |
Fieldmax_completion_tokens | Typeinteger | RequirementOptional | Values & limits Minimum: 1 | Default— |
Fieldn | Typeinteger | RequirementOptional | Values & limits Minimum: 1 | Default— |
Fieldtemperature | Typenumber | RequirementOptional | Values & limits | Default— |
Fieldtop_p | Typenumber | RequirementOptional | Values & limits | Default— |
Fieldseed | Typeinteger | RequirementOptional | Values & limits | Default— |
Fieldstop | Type | RequirementOptional | Values & limits | Default— |
Fieldfrequency_penalty | Typenumber | RequirementOptional | Values & limits | Default— |
Fieldpresence_penalty | Typenumber | RequirementOptional | Values & limits | Default— |
Fielduser | Typestring | RequirementOptional | Values & limits | Default— |
Fieldtools | Typearray | RequirementOptional | Values & limits | Default— |
Fieldservice_tier | Typestring | RequirementOptional | Values & limits , auto, default | Default— |
Fieldmodalities | Typearray | RequirementOptional | Values & limits | Default— |
| Parameter | Contract |
|---|---|
Parametercontent_type | Contractapplication/json |
Parameterrequired | Contractmodel, messages |
Parameteroptional | Contractstream, max_tokens, max_completion_tokens, n, temperature, top_p, seed, stop, frequency_penalty, presence_penalty, user, tools, service_tier, modalities |
Parametermaterial_types | Contracttext |
Parameterconditions | ContractUse text input and text output. Explicit cache, hosted tools and media are unavailable. Standard function tools remain a protocol feature and run in the client. |
Lifecycle: synchronous_json_or_sse
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 / 1K tokens Output · Output tokens $0 / 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}"\nexport AILOOT_API_KEY\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np=json.loads('{"model": "baidu/ernie-4.5-0.3b", "messages": [{"role": "user", "content": "Summarize the supplied text in three concise points."}], "stream": false}')\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\n# Submit once. Retain the complete result, original identifiers and every response header.\nJSON or supported SSE
Keep every complete returned response, file and content part.
Retain the complete response, all actual results, original identifiers, usage and response headers.
Preserve the original response, headers and X-Client-Request-ID. An uncertain outcome does not authorize another creation.
Ailoot Ultra baidu/ernie-4.5-0.3b
Use the method paths shown above.
baidu/ernie-4.5-0.3bERNIE 4.5 0.3B is a text-conversation model in the ERNIE 4.5 open-source family. Use the exact ID baidu/ernie-4.5-0.3b through Chat Completions or Messages to work with supplied text: draft a reply, summarize notes, extract named information, or explain a short passage. Its published profile lists a 120,000-token context and an 8,000-token completion limit; the two values describe different limits.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
ERNIE 4.5 0.3B accepts text conversations through Ailoot Ultra. The exact model ID is baidu/ernie-4.5-0.3b. Put the source material directly in your messages and describe the result you need. This text profile suits tasks such as turning customer notes into a response, condensing a document excerpt, or extracting a specified set of facts into an agreed layout.
A useful request combines the material, the task and the answer requirements. For a support reply, supply the customer message and the policy text the response must follow. For a summary, name the intended reader and ask for decisions, outstanding questions and action items. For extraction, specify the field names and tell the model how to represent facts absent from the source. These instructions give you a result that can be checked against the material you provided.
With Chat Completions, send POST /v1/chat/completions with model and a nonempty messages array. Each message supplies a role and text content. Include the relevant previous turns when the answer depends on a conversation. To work from a document, read its actual text and place the relevant contents in the message; a filename alone does not supply the document.
With Messages, send POST /v1/messages with model, messages and max_tokens. Use user and assistant turns, with an optional separate system instruction. Content can be a text string or text blocks. max_tokens is an integer of at least 1. The published model profile lists an 8,000-token completion ceiling and a 120,000-token context; leave room for the requested answer when preparing the conversation. The context size does not mean an answer can contain 120,000 tokens.
Both interfaces provide the boolean stream control, whose default is false. For a complete JSON answer, retain every returned result field and usage record. For streamed delivery, collect the events and assemble all answer content before treating the response as complete. Chat Completions places the ordinary reply in choices[].message.content; Messages returns content blocks together with stop_reason and usage. Keep the original request identity and any response received if the connection ends before completion, so an incomplete response can be investigated without blindly sending the same request again.
POST /v1/messagestexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits baidu/ernie-4.5-0.3b | Default— |
Fieldmessages | Typearray | RequirementRequired | Values & limits | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits | Defaultfalse |
Fieldsystem | Typestring | RequirementOptional | Values & limits | Default— |
Fieldmax_tokens | Typeinteger | RequirementRequired | Values & limits Minimum: 1 | Default— |
Fieldtemperature | Typenumber | RequirementOptional | Values & limits | Default— |
Fieldtop_p | Typenumber | RequirementOptional | Values & limits | Default— |
Fieldtop_k | Typeinteger | RequirementOptional | Values & limits | Default— |
Fieldstop_sequences | Typearray | RequirementOptional | Values & limits | Default— |
Fieldmetadata | Typeobject | RequirementOptional | Values & limits | Default— |
Fieldtools | Typearray | RequirementOptional | Values & limits | Default— |
Fieldservice_tier | Typestring | RequirementOptional | Values & limits , auto, default | Default— |
Fieldmodalities | Typearray | RequirementOptional | Values & limits | Default— |
| Parameter | Contract |
|---|---|
Parametercontent_type | Contractapplication/json |
Parameterrequired | Contractmodel, messages, max_tokens |
Parameteroptional | Contractstream, system, temperature, top_p, top_k, stop_sequences, metadata, tools, service_tier, modalities |
Parametermaterial_types | Contracttext |
Parameterconditions | ContractUse text input and text output. Explicit cache, hosted tools and media are unavailable. Standard function tools remain a protocol feature and run in the client. |
Lifecycle: synchronous_json_or_sse
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 / 1K tokens Output · Output tokens $0 / 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}"\nexport AILOOT_API_KEY\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np=json.loads('{"model": "baidu/ernie-4.5-0.3b", "messages": [{"role": "user", "content": "Summarize the supplied text in three concise points."}], "stream": false, "max_tokens": 256}')\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/messages \\\n -H "Authorization: Bearer $AILOOT_API_KEY" -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o result.json\n# Submit once. Retain the complete result, original identifiers and every response header.\nJSON or supported SSE
Keep every complete returned response, file and content part.
Retain the complete response, all actual results, original identifiers, usage and response headers.
Preserve the original response, headers and X-Client-Request-ID. An uncertain outcome does not authorize another creation.
Ailoot Ultra baidu/ernie-4.5-0.3b
Use the method paths shown above.