Llama 3.2 1B Instruct is Meta's compact instruction-tuned text model, with a parameter count of 1.23 billion in the maker's table. It accepts multilingual text and returns text or code. The release is intended for dialogue, knowledge retrieval, summarization, mobile writing assistants and rewriting queries or prompts, rather than the image-understanding tasks of Llama's separate vision editions.
Model information and context
The model uses an autoregressive Transformer architecture with grouped-query attention and shared embeddings. Meta distinguishes pretrained and instruction-tuned 1B and 3B models. This entry is the 1B Instruct checkpoint, whose task is to follow the user's request in a conversation rather than serve as the pretrained foundation.
The ordinary text release is documented with a 128k context. The maker separately lists an 8k context for its quantized text models. Those configurations should remain distinct: quantization is not permission to transfer the ordinary release's context to every smaller-format checkpoint or to infer the configured capacity of a particular service.
Use the context for the passages and conversation relevant to one defined operation. A long document can still call for a brief summary, a rewritten search query or a source-based answer. Identify the material the result must retain and the length or fields the next application step needs.
Retrieval, summaries and rewriting
For summarization, provide the actual passage or conversation and specify the audience, length and essential facts. A useful result can separate the central issue, decisions already stated and open questions. If the source contains several topics, identify which topic the summary should cover rather than asking for an undirected reduction of everything.
For knowledge retrieval, supply the selected source passages or information returned by a retrieval component. Ask for the statement answering the question and the passage supporting it. The model's text response can explain retrieved information, while searching a private collection remains work for the connected application. It does not gain that collection merely from the retrieval use-case label.
For query rewriting, give the original question and the purpose of the search. Ask for a concise reformulation that preserves the entity, condition and desired information. A support query can be rewritten with the product term and action that matter, without adding a fact not present in the user's message. Return the revised query separately from any explanation if another component will use it directly.
For a writing assistant, specify the reader, tone and source facts before requesting a short letter or revised passage. For an extraction task, define the fields and how to represent absent information. These are concrete operations that connect the compact model's documented text focus to an artifact that can be inspected.
Supported languages and training
The officially supported languages are English, German, French, Italian, Portuguese, Hindi, Spanish and Thai. The model is trained on a broader language collection, which is distinct from the eight languages explicitly supported by the card. Identify the final language and any terms or identifiers that must remain unchanged.
The family is pretrained on up to nine trillion tokens, with a data cutoff of December 2023. For the 1B and 3B models, the maker uses outputs from Llama 3.1 8B and 70B as token-level training targets and knowledge distillation after pruning. This is a training process; it does not mean the deployed 1B model asks those larger models to answer each user question.
The card describes supervised fine-tuning and preference-based alignment, with post-training rounds involving supervised examples, rejection sampling and direct preference optimization. Those stages produce the Instruct behavior while retaining the named checkpoint and text-only modality.
Conversation and quantized editions
The maker demonstrates separate system and user messages using the matching conversation format. Define the assistant's role and response requirements separately from the actual source text. Keep the relevant earlier answer when a follow-up asks for a revision, clarification or different form of the same information.
The release also documents distinct quantization-aware and post-training quantized variants. Their precision and 8k context belong to those variants rather than a universal specification for this ordinary Instruct entry. Use the exact checkpoint and configuration when comparing a repeated task or preparing a derived deployment.
Substantive boundaries
This 1B release is text-only. It should not inherit image input from Llama 3.2's 11B or 90B vision models, Llama 4's text-and-image capabilities or Muse Spark's named reasoning-effort controls. The card's 256-token generation example is an example setting, not an independently stated maximum answer capacity.
A practical final result is the requested summary, query, extracted record or rewritten text, preserving the source's relevant meaning and identifiers. The read sources do not state a separate hard output-token maximum. Keep the task focused, the input material complete for that operation and the ordinary-versus-quantized context distinction explicit.
", "aboutText": "Llama 3.2 1B Instruct is Meta's compact instruction-tuned text model, with a parameter count of 1.23 billion in the maker's table. It accepts multilingual text and returns text or code. The release is intended for dialogue, knowledge retrieval, summarization, mobile writing assistants and rewriting queries or prompts, rather than the image-understanding tasks of Llama's separate vision editions.\n\nModel information and context\n\nThe model uses an autoregressive Transformer architecture with grouped-query attention and shared embeddings. Meta distinguishes pretrained and instruction-tuned 1B and 3B models. This entry is the 1B Instruct checkpoint, whose task is to follow the user's request in a conversation rather than serve as the pretrained foundation.\n\nThe ordinary text release is documented with a 128k context. The maker separately lists an 8k context for its quantized text models. Those configurations should remain distinct: quantization is not permission to transfer the ordinary release's context to every smaller-format checkpoint or to infer the configured capacity of a particular service.\n\nUse the context for the passages and conversation relevant to one defined operation. A long document can still call for a brief summary, a rewritten search query or a source-based answer. Identify the material the result must retain and the length or fields the next application step needs.\n\nRetrieval, summaries and rewriting\n\nFor summarization, provide the actual passage or conversation and specify the audience, length and essential facts. A useful result can separate the central issue, decisions already stated and open questions. If the source contains several topics, identify which topic the summary should cover rather than asking for an undirected reduction of everything.\n\nFor knowledge retrieval, supply the selected source passages or information returned by a retrieval component. Ask for the statement answering the question and the passage supporting it. The model's text response can explain retrieved information, while searching a private collection remains work for the connected application. It does not gain that collection merely from the retrieval use-case label.\n\nFor query rewriting, give the original question and the purpose of the search. Ask for a concise reformulation that preserves the entity, condition and desired information. A support query can be rewritten with the product term and action that matter, without adding a fact not present in the user's message. Return the revised query separately from any explanation if another component will use it directly.\n\nFor a writing assistant, specify the reader, tone and source facts before requesting a short letter or revised passage. For an extraction task, define the fields and how to represent absent information. These are concrete operations that connect the compact model's documented text focus to an artifact that can be inspected.\n\nSupported languages and training\n\nThe officially supported languages are English, German, French, Italian, Portuguese, Hindi, Spanish and Thai. The model is trained on a broader language collection, which is distinct from the eight languages explicitly supported by the card. Identify the final language and any terms or identifiers that must remain unchanged.\n\nThe family is pretrained on up to nine trillion tokens, with a data cutoff of December 2023. For the 1B and 3B models, the maker uses outputs from Llama 3.1 8B and 70B as token-level training targets and knowledge distillation after pruning. This is a training process; it does not mean the deployed 1B model asks those larger models to answer each user question.\n\nThe card describes supervised fine-tuning and preference-based alignment, with post-training rounds involving supervised examples, rejection sampling and direct preference optimization. Those stages produce the Instruct behavior while retaining the named checkpoint and text-only modality.\n\nConversation and quantized editions\n\nThe maker demonstrates separate system and user messages using the matching conversation format. Define the assistant's role and response requirements separately from the actual source text. Keep the relevant earlier answer when a follow-up asks for a revision, clarification or different form of the same information.\n\nThe release also documents distinct quantization-aware and post-training quantized variants. Their precision and 8k context belong to those variants rather than a universal specification for this ordinary Instruct entry. Use the exact checkpoint and configuration when comparing a repeated task or preparing a derived deployment.\n\nSubstantive boundaries\n\nThis 1B release is text-only. It should not inherit image input from Llama 3.2's 11B or 90B vision models, Llama 4's text-and-image capabilities or Muse Spark's named reasoning-effort controls. The card's 256-token generation example is an example setting, not an independently stated maximum answer capacity.\n\nA practical final result is the requested summary, query, extracted record or rewritten text, preserving the source's relevant meaning and identifiers. The read sources do not state a separate hard output-token maximum. Keep the task focused, the input material complete for that operation and the ordinary-versus-quantized context distinction explicit.", "editorialOverlay": { "copy_fields": { "practical_html": "Technical example in English
Rewrite a retrieval query\n\nRewrite this customer question as one concise search query for a support knowledge base: “My Atlas Mini still shows the old timezone after I changed it in account settings. Do I need to restart?” Preserve the product, stale timezone symptom and restart question. Return only the query and do not invent a cause.A single concise retrieval query retaining the product name, symptom and requested information.Check that the query includes Atlas Mini and timezone refresh or restart, without asserting a confirmed defect.meta-llama/llama-3.2-1b-instructLlama 3.2 1B Instruct is Meta's compact 1.23B text model for dialogue, summarization, knowledge retrieval and query rewriting. The maker lists a 128k context for the ordinary release and 8k for separate quantized variants, with multilingual text and code output.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Llama 3.2 1B Instruct is Meta's compact instruction-tuned text model, with a parameter count of 1.23 billion in the maker's table. It accepts multilingual text and returns text or code. The release is intended for dialogue, knowledge retrieval, summarization, mobile writing assistants and rewriting queries or prompts, rather than the image-understanding tasks of Llama's separate vision editions.
Model information and context
The model uses an autoregressive Transformer architecture with grouped-query attention and shared embeddings. Meta distinguishes pretrained and instruction-tuned 1B and 3B models. This entry is the 1B Instruct checkpoint, whose task is to follow the user's request in a conversation rather than serve as the pretrained foundation.
The ordinary text release is documented with a 128k context. The maker separately lists an 8k context for its quantized text models. Those configurations should remain distinct: quantization is not permission to transfer the ordinary release's context to every smaller-format checkpoint or to infer the configured capacity of a particular service.
Use the context for the passages and conversation relevant to one defined operation. A long document can still call for a brief summary, a rewritten search query or a source-based answer. Identify the material the result must retain and the length or fields the next application step needs.
Retrieval, summaries and rewriting
For summarization, provide the actual passage or conversation and specify the audience, length and essential facts. A useful result can separate the central issue, decisions already stated and open questions. If the source contains several topics, identify which topic the summary should cover rather than asking for an undirected reduction of everything.
For knowledge retrieval, supply the selected source passages or information returned by a retrieval component. Ask for the statement answering the question and the passage supporting it. The model's text response can explain retrieved information, while searching a private collection remains work for the connected application. It does not gain that collection merely from the retrieval use-case label.
For query rewriting, give the original question and the purpose of the search. Ask for a concise reformulation that preserves the entity, condition and desired information. A support query can be rewritten with the product term and action that matter, without adding a fact not present in the user's message. Return the revised query separately from any explanation if another component will use it directly.
For a writing assistant, specify the reader, tone and source facts before requesting a short letter or revised passage. For an extraction task, define the fields and how to represent absent information. These are concrete operations that connect the compact model's documented text focus to an artifact that can be inspected.
Supported languages and training
The officially supported languages are English, German, French, Italian, Portuguese, Hindi, Spanish and Thai. The model is trained on a broader language collection, which is distinct from the eight languages explicitly supported by the card. Identify the final language and any terms or identifiers that must remain unchanged.
The family is pretrained on up to nine trillion tokens, with a data cutoff of December 2023. For the 1B and 3B models, the maker uses outputs from Llama 3.1 8B and 70B as token-level training targets and knowledge distillation after pruning. This is a training process; it does not mean the deployed 1B model asks those larger models to answer each user question.
The card describes supervised fine-tuning and preference-based alignment, with post-training rounds involving supervised examples, rejection sampling and direct preference optimization. Those stages produce the Instruct behavior while retaining the named checkpoint and text-only modality.
Conversation and quantized editions
The maker demonstrates separate system and user messages using the matching conversation format. Define the assistant's role and response requirements separately from the actual source text. Keep the relevant earlier answer when a follow-up asks for a revision, clarification or different form of the same information.
The release also documents distinct quantization-aware and post-training quantized variants. Their precision and 8k context belong to those variants rather than a universal specification for this ordinary Instruct entry. Use the exact checkpoint and configuration when comparing a repeated task or preparing a derived deployment.
Substantive boundaries
This 1B release is text-only. It should not inherit image input from Llama 3.2's 11B or 90B vision models, Llama 4's text-and-image capabilities or Muse Spark's named reasoning-effort controls. The card's 256-token generation example is an example setting, not an independently stated maximum answer capacity.
A practical final result is the requested summary, query, extracted record or rewritten text, preserving the source's relevant meaning and identifiers. The read sources do not state a separate hard output-token maximum. Keep the task focused, the input material complete for that operation and the ordinary-versus-quantized context distinction explicit.
POST /v1/chat/completionstexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits meta-llama/llama-3.2-1b-instruct | 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.
Match each charge to the conditions shown. Use one rate for each set of alternative configurations.
| Configuration | Rate schedule | Context tokens, more than | Context tokens, up to and including | Charge |
|---|---|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Context tokens, more than— | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.000054 / 1K tokens Output · Output tokens $0.000402 / 1K tokens Input · Cache read — / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
| ConfigurationRequest-length tier 1 | Rate schedule— | Context tokens, more than0 | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.000054 / 1K tokens Output · Output tokens $0.000402 / 1K tokens Input · Cache read — / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np=json.loads('{"model": "meta-llama/llama-3.2-1b-instruct", "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='meta-llama/llama-3.2-1b-instruct',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='meta-llama/llama-3.2-1b-instruct',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 meta-llama/llama-3.2-1b-instruct
Use the method paths shown above.
meta-llama/llama-3.2-1b-instructLlama 3.2 1B Instruct is Meta's compact 1.23B text model for dialogue, summarization, knowledge retrieval and query rewriting. The maker lists a 128k context for the ordinary release and 8k for separate quantized variants, with multilingual text and code output.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Llama 3.2 1B Instruct is Meta's compact instruction-tuned text model, with a parameter count of 1.23 billion in the maker's table. It accepts multilingual text and returns text or code. The release is intended for dialogue, knowledge retrieval, summarization, mobile writing assistants and rewriting queries or prompts, rather than the image-understanding tasks of Llama's separate vision editions.
Model information and context
The model uses an autoregressive Transformer architecture with grouped-query attention and shared embeddings. Meta distinguishes pretrained and instruction-tuned 1B and 3B models. This entry is the 1B Instruct checkpoint, whose task is to follow the user's request in a conversation rather than serve as the pretrained foundation.
The ordinary text release is documented with a 128k context. The maker separately lists an 8k context for its quantized text models. Those configurations should remain distinct: quantization is not permission to transfer the ordinary release's context to every smaller-format checkpoint or to infer the configured capacity of a particular service.
Use the context for the passages and conversation relevant to one defined operation. A long document can still call for a brief summary, a rewritten search query or a source-based answer. Identify the material the result must retain and the length or fields the next application step needs.
Retrieval, summaries and rewriting
For summarization, provide the actual passage or conversation and specify the audience, length and essential facts. A useful result can separate the central issue, decisions already stated and open questions. If the source contains several topics, identify which topic the summary should cover rather than asking for an undirected reduction of everything.
For knowledge retrieval, supply the selected source passages or information returned by a retrieval component. Ask for the statement answering the question and the passage supporting it. The model's text response can explain retrieved information, while searching a private collection remains work for the connected application. It does not gain that collection merely from the retrieval use-case label.
For query rewriting, give the original question and the purpose of the search. Ask for a concise reformulation that preserves the entity, condition and desired information. A support query can be rewritten with the product term and action that matter, without adding a fact not present in the user's message. Return the revised query separately from any explanation if another component will use it directly.
For a writing assistant, specify the reader, tone and source facts before requesting a short letter or revised passage. For an extraction task, define the fields and how to represent absent information. These are concrete operations that connect the compact model's documented text focus to an artifact that can be inspected.
Supported languages and training
The officially supported languages are English, German, French, Italian, Portuguese, Hindi, Spanish and Thai. The model is trained on a broader language collection, which is distinct from the eight languages explicitly supported by the card. Identify the final language and any terms or identifiers that must remain unchanged.
The family is pretrained on up to nine trillion tokens, with a data cutoff of December 2023. For the 1B and 3B models, the maker uses outputs from Llama 3.1 8B and 70B as token-level training targets and knowledge distillation after pruning. This is a training process; it does not mean the deployed 1B model asks those larger models to answer each user question.
The card describes supervised fine-tuning and preference-based alignment, with post-training rounds involving supervised examples, rejection sampling and direct preference optimization. Those stages produce the Instruct behavior while retaining the named checkpoint and text-only modality.
Conversation and quantized editions
The maker demonstrates separate system and user messages using the matching conversation format. Define the assistant's role and response requirements separately from the actual source text. Keep the relevant earlier answer when a follow-up asks for a revision, clarification or different form of the same information.
The release also documents distinct quantization-aware and post-training quantized variants. Their precision and 8k context belong to those variants rather than a universal specification for this ordinary Instruct entry. Use the exact checkpoint and configuration when comparing a repeated task or preparing a derived deployment.
Substantive boundaries
This 1B release is text-only. It should not inherit image input from Llama 3.2's 11B or 90B vision models, Llama 4's text-and-image capabilities or Muse Spark's named reasoning-effort controls. The card's 256-token generation example is an example setting, not an independently stated maximum answer capacity.
A practical final result is the requested summary, query, extracted record or rewritten text, preserving the source's relevant meaning and identifiers. The read sources do not state a separate hard output-token maximum. Keep the task focused, the input material complete for that operation and the ordinary-versus-quantized context distinction explicit.
POST /v1/messagestextjson, sse| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits meta-llama/llama-3.2-1b-instruct | 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.
Match each charge to the conditions shown. Use one rate for each set of alternative configurations.
| Configuration | Rate schedule | Context tokens, more than | Context tokens, up to and including | Charge |
|---|---|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Context tokens, more than— | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.000054 / 1K tokens Output · Output tokens $0.000402 / 1K tokens Input · Cache read — / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
| ConfigurationRequest-length tier 1 | Rate schedule— | Context tokens, more than0 | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.000054 / 1K tokens Output · Output tokens $0.000402 / 1K tokens Input · Cache read — / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
{\n "model": "meta-llama/llama-3.2-1b-instruct",\n "max_tokens": 128,\n "messages": [\n {\n "role": "user",\n "content": "Summarize the incident report."\n }\n ],\n "stream": false\n}set -eu\n: "${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': 'meta-llama/llama-3.2-1b-instruct', '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': 'meta-llama/llama-3.2-1b-instruct', '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 meta-llama/llama-3.2-1b-instruct
Use the method paths shown above.
Technical example in English
Rewrite a retrieval query\n\nRewrite this customer question as one concise search query for a support knowledge base: “My Atlas Mini still shows the old timezone after I changed it in account settings. Do I need to restart?” Preserve the product, stale timezone symptom and restart question. Return only the query and do not invent a cause.A single concise retrieval query retaining the product name, symptom and requested information.Check that the query includes Atlas Mini and timezone refresh or restart, without asserting a confirmed defect.