QwQ

reasoning
try online

QwQ-32B is an innovative language model developed by Alibaba, featuring 32 billion parameters and a context window of 131K tokens. It is specifically designed for deep reasoning and excels at multi-step logical analysis, making it highly effective in solving complex tasks that require structured thinking.

QwQ-32B was trained using advanced reinforcement learning techniques, significantly enhancing its reasoning capabilities. This enables the model to deliver outstanding performance in areas such as mathematical computation, programming, and legal document analysis. In terms of performance, it rivals DeepSeek-R1, which has 671 billion parameters. Additionally, QwQ-32B possesses agent-like behavior capabilities, allowing it to adapt its reasoning based on feedback and utilize various tools for more accurate query analysis.

Thanks to its context window of 131,000 tokens, the model can handle large-scale analytical tasks and work with multi-step logical reasoning chains. This makes it indispensable for scientific research, educational applications, identifying issues in code, comparing arguments in legal documents, and other tasks that demand maximum attention to detail.


Announce Date: 06.03.2025
Parameters: 32.5B
Context: 131K
Layers: 64
Attention Type: Full or Sliding Window Attention
VRAM requirements: 49.6 GB using 4 bits quantization
Developer: Qwen
Transformers Version: 4.43.1
License: Apache 2.0

Public endpoint

Use our pre-built public endpoints for free to test inference and explore QwQ capabilities. You can obtain an API access token on the token management page after registration and verification.
Model Name Context Type GPU TPS Status Link
Qwen/QwQ-32B 40,960.0 Public 2×RTX4090 46.40 AVAILABLE chat

API access to QwQ endpoints

curl https://chat.immers.cloud/v1/endpoints/QwQ-32b/generate/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer USER_API_KEY" \
-d '{"model": "QwQ-32b", "messages": [
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Say this is a test"}
], "temperature": 0, "max_tokens": 150}'
$response = Invoke-WebRequest https://chat.immers.cloud/v1/endpoints/QwQ-32b/generate/chat/completions `
-Method POST `
-Headers @{
"Authorization" = "Bearer USER_API_KEY"
"Content-Type" = "application/json"
} `
-Body (@{
model = "QwQ-32b"
messages = @(
@{ role = "system"; content = "You are a helpful assistant." },
@{ role = "user"; content = "Say this is a test" }
)
} | ConvertTo-Json)
($response.Content | ConvertFrom-Json).choices[0].message.content
#!pip install OpenAI --upgrade

from openai import OpenAI

client = OpenAI(
api_key="USER_API_KEY",
base_url="https://chat.immers.cloud/v1/endpoints/QwQ-32b/generate/",
)

chat_response = client.chat.completions.create(
model="QwQ-32b",
messages=[
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Say this is a test"},
]
)
print(chat_response.choices[0].message.content)

Private server

Rent your own physically dedicated instance with hourly or long-term monthly billing.

We recommend deploying private instances in the following scenarios:

  • maximize endpoint performance,
  • enable full context for long sequences,
  • ensure top-tier security for data processing in an isolated, dedicated environment,
  • use custom weights, such as fine-tuned models or LoRA adapters.

Recommended configurations for hosting QwQ

Prices:
Name vCPU RAM, MB Disk, GB GPU Price, hour
teslat4-4.16.64.160
131,072.0
 16 65536 160 4 $0.96 Launch
teslaa2-4.32.128.160
131,072.0
 32 131072 160 4 $1.26 Launch
teslaa10-3.16.96.160
131,072.0
 16 98304 160 3 $1.34 Launch
teslav100-2.16.64.240
131,072.0
 16 65535 240 2 $2.22 Launch
rtxa5000-4.16.128.160.nvlink
131,072.0
 16 131072 160 4 $2.34 Launch
rtx3090-3.16.96.160
131,072.0
 16 98304 160 3 $2.45 Launch
teslaa100-1.16.64.160
131,072.0
 16 65536 160 1 $2.58 Launch
rtx5090-2.16.64.160
131,072.0
 16 65536 160 2 $2.93 Launch
rtx4090-3.16.96.160
131,072.0
 16 98304 160 3 $3.23 Launch
teslah100-1.16.64.160
131,072.0
 16 65536 160 1 $5.11 Launch
h200-1.16.128.160
131,072.0
 16 131072 160 1 $6.98 Launch
Prices:
Name vCPU RAM, MB Disk, GB GPU Price, hour
teslaa2-6.32.128.160
131,072.0
 32 131072 160 6 $1.65 Launch
teslaa10-4.16.128.160
131,072.0
 16 131072 160 4 $1.75 Launch
rtxa5000-4.16.128.160.nvlink
131,072.0
 16 131072 160 4 $2.34 Launch
teslaa100-1.16.128.160
131,072.0
 16 131072 160 1 $2.71 Launch
rtx3090-4.16.96.320
131,072.0
 16 98304 320 4 $3.18 Launch
teslav100-3.64.256.320
131,072.0
 64 262144 320 3 $3.89 Launch
rtx4090-4.16.96.320
131,072.0
 16 98304 320 4 $4.22 Launch
rtx5090-3.16.96.160
131,072.0
 16 98304 160 3 $4.34 Launch
teslah100-1.16.128.160
131,072.0
 16 131072 160 1 $5.23 Launch
h200-1.16.128.160
131,072.0
 16 131072 160 1 $6.98 Launch
Prices:
Name vCPU RAM, MB Disk, GB GPU Price, hour
rtxa5000-6.24.192.160.nvlink
131,072.0
 24 196608 160 6 $3.50 Launch
teslav100-4.32.256.160
131,072.0
 32 262144 160 4 $4.66 Launch
teslaa100-2.24.128.160.nvlink
131,072.0
 24 131072 160 2 $5.10 Launch
rtx5090-4.16.128.160
131,072.0
 16 131072 160 4 $5.74 Launch
rtx4090-6.44.256.160
131,072.0
 44 262144 160 6 $6.63 Launch
h200-1.16.128.160
131,072.0
 16 131072 160 1 $6.98 Launch
teslah100-2.24.256.160
131,072.0
 24 262144 160 2 $10.40 Launch

Related models

Qwen2-0.5B
Qwen2-1.5B
Qwen2-7B
Qwen2-72B
Qwen2.5-0.5B
Qwen2.5-1.5B
Qwen2.5-3B
Qwen2.5-7B
Qwen2.5-14B
Qwen2.5-32B
Qwen2.5-72B
Qwen2-57B-A14B
DeepSeek-R1-Distill-Qwen-32B
DeepSeek-R1-Distill-Qwen-1.5B
Qwen2.5-VL-3B
Qwen2.5-VL-7B
Qwen2.5-7B-1M

Need help?

Contact our dedicated neural networks support team at nn@immers.cloud or send your request to the sales department at sale@immers.cloud.