Qwen: Qwen2.5 VL 72B Instruct

qwen/qwen2.5-vl-72b-instruct

Qwen2.5-VL is proficient in recognizing common objects such as flowers, birds, fish, and insects. It is also highly capable of analyzing texts, charts, icons, graphics, and layouts within images.

vision

MODALITIES

→

INPUT PRICE

$0.80per 1M

OUTPUT PRICE

$1per 1M

CONTEXT

128K

RELEASED

Feb 1, 2025

Provider					Chat
FastAPI	$0.8	$1	Hit rate— Cache read$0.4	99.0%

Capabilities

Input modalities

imagetext

Output modalities

text

Features

frequency_penaltylogit_biaslogprobsmax_tokenspresence_penaltyrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetop_ktop_logprobstop_p

Get API Key

Create an API key from the Keys page, then set it as an environment variable:

export ONLIST_API_KEY=sk-...

Make your first request

Endpoints

POSThttps://onlist.io/v1/chat/completions

OpenAI Chat Completions format

Request Headers

Authorization:Bearer $ONLIST_API_KEY

Content-Type:application/json

Model:qwen/qwen2.5-vl-72b-instruct

POSThttps://onlist.io/v1/responses

OpenAI Responses format

Request Headers

Authorization:Bearer $ONLIST_API_KEY

Content-Type:application/json

Model:qwen/qwen2.5-vl-72b-instruct

POSThttps://onlist.io/v1/messages

Anthropic Messages format

Request Headers

Authorization:Bearer $ONLIST_API_KEY

Content-Type:application/json

Model:qwen/qwen2.5-vl-72b-instruct

Code samples

curl https://onlist.io/v1/chat/completions \
  -H "Authorization: Bearer $ONLIST_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
       "model": "qwen/qwen2.5-vl-72b-instruct",
       "messages": [
         {
           "role": "user",
           "content": "Explain quantum entanglement in one paragraph."
         }
       ]
     }'

Replace $ONLIST_API_KEY with the API key from your Keys page.

Authentication

All requests must include an Authorization: Bearer <TOKEN> header. Generate keys from the Keys page; keys can be scoped to specific models, groups, IP ranges, and rate limits.

Enable streaming

Add "stream": true to receive partial responses as server-sent events in real time.

Streaming example

curl https://onlist.io/v1/chat/completions \
  -H "Authorization: Bearer $ONLIST_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
       "model": "qwen/qwen2.5-vl-72b-instruct",
       "messages": [
         {
           "role": "user",
           "content": "Write a haiku about recursion."
         }
       ],
       "stream": true
     }'

Supported parameters

Name	Type	Description
`frequency_penalty`	number	Penalizes new tokens based on their existing frequency in the text (-2.0 to 2.0).
`logit_bias`
`logprobs`	boolean	Whether to return log probabilities of the output tokens.
`max_tokens`	integer	Maximum number of tokens to generate in the completion.
`presence_penalty`	number	Penalizes new tokens based on whether they appear in the text so far (-2.0 to 2.0).
`repetition_penalty`	number	Penalizes repeated tokens. Values > 1 discourage repetition.
`response_format`	object	Specifies the output format. Use {"type": "json_object"} for JSON mode.
`seed`	integer	If specified, the system will attempt deterministic sampling for reproducible results.
`stop`	string \| array	Up to 4 sequences where the API will stop generating tokens.
`structured_outputs`
`temperature`	number	Sampling temperature between 0 and 2. Higher values make output more random.
`top_k`	integer	Limits token selection to the k most likely candidates at each step.
`top_logprobs`	integer	Number of most likely tokens to return at each position (0-20). Requires logprobs: true.
`top_p`	number	Nucleus sampling. The model considers tokens with top_p probability mass.

These are the request parameters this model accepts. Parameter semantics follow the OpenAI Chat Completions specification.