Qwen2.5-Coder-3B
Qwen2.5-Coder-3B-Instruct is Qwen’s compact instruction-tuned model for code generation, reasoning and repair. Its exact 3.09B-parameter profile has a 32,768-token context for coding briefs that include the relevant implementation, interface and expected behaviour.
- Ailoot Ultra · OpenAI Chat
POST /v1/chat/completionsqwen2.5-coder-3b-instruct - Ailoot Ultra · Anthropic Messages
POST /v1/messagesqwen2.5-coder-3b-instruct
About the model
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Qwen2.5-Coder-3B-Instruct is Qwen’s compact instruction-tuned coding model, with 3.09 billion total parameters and a full context of 32,768 tokens. The producer describes code generation, code reasoning and code fixing as core family purposes, while retaining mathematical and general-language abilities. The family’s training uses 5.5 trillion tokens of code, text–code grounding and synthetic material.
Prepare a focused coding brief
Supply the language, function signature, input assumptions and expected output before requesting code. Include the example that establishes the intended behaviour, particularly when a boundary case changes the implementation. For a repair, provide the current function and the failure it should address.
An illustrative request can ask for a parser that retains the approved field names and returns the specified result for an empty input. Keep the relevant interface and examples with the proposed code so the review concerns the same contract.
Use the Instruct conversation template
This checkpoint is the Instruct edition, separate from the Base model. The native example supplies system and user messages through the checkpoint’s chat template and decodes the generated continuation as text. Keep the task instruction and current code question clear when preparing that conversation.
For an explanation, identify the function or branch to analyse and ask which assumption determines its behaviour. For a rewrite, state which interfaces and output fields should remain unchanged. A useful response can present the proposed code followed by the reasoning connecting the change to those requirements.
Fit the relevant dependencies into the 3B profile
The exact 3B card publishes a 32,768-token context, with 2.77 billion non-embedding parameters and 36 layers. Its model architecture uses grouped-query attention, with 16 query heads and two key/value heads. Select the relevant project material for the request rather than treating this compact edition as the separately configured 32B model.
For an illustrative cross-function task, include the implementation, its caller and the type or schema that connects them. Preserve file paths and names so the generated change refers to the intended location. If a question concerns only one condition, retain the code that establishes it and the examples that determine the expected result.
Produce an artifact that can be checked
Choose whether the next step needs a complete function, a targeted code change or a written explanation. The native generation example uses an illustrative output allowance for a small task; define the actual deliverable and a suitable configured response allowance for the work at hand.
Keep the source code, requirement and proposed change together when the surrounding application runs or checks the result. For a correction, supply the actual failure and current implementation, then request the change that addresses that evidence. Retain the selected 3B Instruct name with the completed code and explanation so the continuation uses the same producer edition.
Through Ailoot, select this exact model ID in /v1/chat/completions and supply the task in messages. Each message carries a role and its content; keep the relevant instruction and source material together. Collect the complete JSON response or all SSE segments when stream is enabled, and retain usage with the written result.
Ailoot Ultra · OpenAI Chat POST /v1/chat/completions qwen2.5-coder-3b-instruct
qwen2.5-coder-3b-instructOverview
Qwen2.5-Coder-3B-Instruct is Qwen’s compact instruction-tuned model for code generation, reasoning and repair. Its exact 3.09B-parameter profile has a 32,768-token context for coding briefs that include the relevant implementation, interface and expected behaviour.
- Accepted inputs
- Text · Images
- Results
- Text
Available in Ailoot · inputs & methods
Ailoot Ultra · OpenAI Chat POST /v1/chat/completions
- Input types
text,image- Output types
text- Output format
- JSON or SSE
| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits qwen2.5-coder-3b-instruct Conditions: Use this exact Ailoot routing selector. | Default— |
Fieldmessages | Typearray | RequirementRequired | Values & limits Conditions: | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits Conditions: | Defaultfalse |
Input material combinations
text_input · Required: messages
image_with_text · Required: messages with actual image content blocks
Method parameters and lifecycle
| Parameter | Contract |
|---|---|
Parametercontent_type | Contractapplication/json |
Parameterrequired | Contractmodel, messages |
Parameteroptional | Contractstream |
Parametermaterial_types | Contracttext, image |
Parameterconditions | ContractUse this exact Ailoot routing selector., Image content is carried by this protocol where the exact selected model supports vision. Protocol admission does not establish image capability for every model. |
Method aliases: /chat/completions
Lifecycle: synchronous_or_streamed
Pricing
Tariff references do not add request options. Supported parameters are listed for the selected method.
- Fast mode
- Standard rate When fast mode is selected
- Flex mode
- Standard rate When flex mode is selected
- Maximum reasoning effort
- Standard rate When maximum reasoning effort uses the default adjustment
- Explicit cache pricing
- Not enabled When explicit cache is requested
- Text-only zero rates
- Cache, tools and media follow their separately listed pricing conditions. For text-only input and output
Ailoot rates
Prices by configuration
| 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 $0 / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
Example requests
Ailoot Ultra · OpenAI Chat
set -eu
: "${AILOOT_API_KEY:?Set your Ailoot API key}"
export AILOOT_API_KEY
python3 - <<'PY_REQUEST'
import json,os
from pathlib import Path
p=json.loads('{"model": "qwen2.5-coder-3b-instruct", "messages": [{"role": "user", "content": "Explain how a rainbow forms."}]}')
Path("request.json").write_text(json.dumps(p,ensure_ascii=False))
PY_REQUEST
curl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/chat/completions \
-H "Authorization: Bearer $AILOOT_API_KEY" -H 'Content-Type: application/json' \
--data-binary @request.json -D response-headers.txt -o result.json
# Submit once. Retain the complete result, original identifiers and every response header.
Receive and save results
Ailoot Ultra · OpenAI Chat
JSON or SSE
Keep all completed results
Keep every complete returned response, file and content part.
Retain every content block, tool-call ID, thinking signature when returned, all SSE deltas and complete usage. A first text fragment is not the whole result.
If the original response needs checking
Keep the original X-Client-Request-ID, response and headers. An unknown or result_pending_check outcome does not authorize a second creation.
Response fields
| Field | Type |
|---|---|
Fieldchoices[] | Typeobject |
Fieldusage | Typeobject |
Errors and next steps
| Response | Condition | Next step |
|---|---|---|
Response400 · invalid_request_error | ConditionA required field, model selector, type or input representation is invalid. | Next stepInspect the returned validation detail before preparing a separate request. |
Response | ConditionThe original request or result is unknown or requires a check. | Next stepKeep original identifiers and headers; read only supported status/library records. Do not submit creation again. |
Aliases
Model aliases
Ailoot Ultra qwen2.5-coder-3b-instruct
Method path aliases
/chat/completions
Related caller IDs
Ailoot Ultra · Anthropic Messages POST /v1/messages qwen2.5-coder-3b-instruct
qwen2.5-coder-3b-instructOverview
Qwen2.5-Coder-3B-Instruct is Qwen’s compact instruction-tuned model for code generation, reasoning and repair. Its exact 3.09B-parameter profile has a 32,768-token context for coding briefs that include the relevant implementation, interface and expected behaviour.
- Accepted inputs
- Text · Images
- Results
- json · sse
Available in Ailoot · inputs & methods
Ailoot Ultra · Anthropic Messages POST /v1/messages
- Input types
text,image- Output types
json,sse- Output format
- JSON or SSE
| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits qwen2.5-coder-3b-instruct Conditions: Use this exact Ailoot routing selector. | Default— |
Fieldmessages | Typearray | RequirementRequired | Values & limits Conditions: | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits Conditions: | Defaultfalse |
Fieldmax_tokens | Typeinteger | RequirementRequired | Values & limits Minimum: 1 Conditions: | Default— |
Input material combinations
text_input · Required: messages
image_with_text · Required: messages with actual image content blocks
Method parameters and lifecycle
| Parameter | Contract |
|---|---|
Parametercontent_type | Contractapplication/json |
Parameterrequired | Contractmodel, messages, max_tokens |
Parameteroptional | Contractstream |
Parametermaterial_types | Contracttext, image |
Parameterconditions | ContractUse this exact Ailoot routing selector., Image content is carried by this protocol where the exact selected model supports vision. Protocol admission does not establish image capability for every model. |
Lifecycle: synchronous_or_streamed
Pricing
Tariff references do not add request options. Supported parameters are listed for the selected method.
- Fast mode
- Standard rate When fast mode is selected
- Flex mode
- Standard rate When flex mode is selected
- Maximum reasoning effort
- Standard rate When maximum reasoning effort uses the default adjustment
- Explicit cache pricing
- Not enabled When explicit cache is requested
- Text-only zero rates
- Cache, tools and media follow their separately listed pricing conditions. For text-only input and output
Ailoot rates
Prices by configuration
| 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 $0 / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
Example requests
Ailoot Ultra · Anthropic Messages
set -eu
: "${AILOOT_API_KEY:?Set your Ailoot API key}"
export AILOOT_API_KEY
python3 - <<'PY_REQUEST'
import json,os
from pathlib import Path
p=json.loads('{"model": "qwen2.5-coder-3b-instruct", "messages": [{"role": "user", "content": "Explain how a rainbow forms."}], "max_tokens": 256}')
Path("request.json").write_text(json.dumps(p,ensure_ascii=False))
PY_REQUEST
curl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/messages \
-H "Authorization: Bearer $AILOOT_API_KEY" -H 'Content-Type: application/json' \
--data-binary @request.json -D response-headers.txt -o result.json
# Submit once. Retain the complete result, original identifiers and every response header.
Receive and save results
Ailoot Ultra · Anthropic Messages
JSON or SSE
Keep all completed results
Keep every complete returned response, file and content part.
Retain every content block, tool-call ID, thinking signature when returned, all SSE deltas and complete usage. A first text fragment is not the whole result.
If the original response needs checking
Keep the original X-Client-Request-ID, response and headers. An unknown or result_pending_check outcome does not authorize a second creation.
Response fields
| Field | Type |
|---|---|
Fieldcontent[] | Typeobject |
Fieldusage | Typeobject |
Errors and next steps
| Response | Condition | Next step |
|---|---|---|
Response400 · invalid_request_error | ConditionA required field, model selector, type or input representation is invalid. | Next stepInspect the returned validation detail before preparing a separate request. |
Response | ConditionThe original request or result is unknown or requires a check. | Next stepKeep original identifiers and headers; read only supported status/library records. Do not submit creation again. |
Aliases
Model aliases
Ailoot Ultra qwen2.5-coder-3b-instruct
Method path aliases
/chat/completions
Related caller IDs
FAQ · Qwen2.5-Coder-3B
- What inputs and results are available?
POST /v1/chat/completions·qwen2.5-coder-3b-instruct:text,image→textPOST /v1/messages·qwen2.5-coder-3b-instruct:text,image→json,sse- Which fields does the request use?
POST /v1/chat/completions·qwen2.5-coder-3b-instruct:model,messages,streamPOST /v1/messages·qwen2.5-coder-3b-instruct:model,messages,stream,max_tokens- What limits apply?
POST /v1/messages·qwen2.5-coder-3b-instruct:max_tokens.minimum:1