Requesty
DeepInfra Inc. logo
DeepInfra Inc.

Qwen/Qwen2.5-72B-Instruct

Qwen3, the latest generation in the Qwen large language model series, features both dense and mixture-of-experts (MoE) architectures to excel in reasoning, multilingual support, and advanced agent tasks. Its unique ability to switch seamlessly between a thinking mode for complex reasoning and a non-thinking mode for efficient dialogue ensures versatile, high-quality performance. Significantly outperforming prior models like QwQ and Qwen2.5, Qwen3 delivers superior mathematics, coding, commonsense reasoning, creative writing, and interactive dialogue capabilities. The Qwen3-30B-A3B variant includes 30.5 billion parameters (3.3 billion activated), 48 layers, 128 experts (8 activated per task), and supports up to 131K token contexts with YaRN, setting a new standard among open-source models.

Tool callingJSON schema
Try Qwen/Qwen2.5-72B-InstructCompareDocs

Specifications

Context window131K tokens
Max outputβ€”
API typechat
AddedApr 28, 2025
Model IDdeepinfra/Qwen/Qwen2.5-72B-Instruct
Data retentionNo
Used for trainingNo
Provider locationπŸ‡ΊπŸ‡Έ US
Privacy policyDeepInfra Privacy Policy→

Benchmarks

Released 2024-09-19
Coding Indexcoding
11.9%

Artificial Analysis Coding Index β€” a composite of coding evaluations including LiveCodeBench, SciCode and Terminal-Bench.

GPQA Diamondreasoning
49.1%

Graduate-level physics, chemistry & biology questions designed to resist Googling.

Intelligence Indexreasoning
15.6%

Artificial Analysis Intelligence Index β€” a composite of multiple evaluations measuring overall model capability.

Scores are sourced from official model cards, Artificial Analysis, and public leaderboards. Benchmarks measure specific skills and do not capture every aspect of model quality β€” always test on your own workload.

Pricing

Input / 1M
$0.23
Output / 1M
$0.40
Cache write
β€”
Cache read
β€”
Estimated cost
100K input + 10K output$0.0270
1M input + 100K output$0.27
10M input + 1M output$2.70

Requesty charges exactly what the upstream provider charges β€” no markup, no per-request fees. Prompt caching and smart routing can reduce effective cost by 30-80%.

Quickstart

Drop-in compatible with the OpenAI SDK. Change the base URL, swap in your Requesty API key, and set the model to deepinfra/Qwen/Qwen2.5-72B-Instruct.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_REQUESTY_API_KEY",
    base_url="https://router.requesty.ai/v1",
)

response = client.chat.completions.create(
    model="deepinfra/Qwen/Qwen2.5-72B-Instruct",
    messages=[
        {"role": "user", "content": "Explain quantum computing in one paragraph."},
    ],
)

print(response.choices[0].message.content)

Other DeepInfra Inc. models

zai-org/GLM-5.1
203K ctx$1.05/1M
Coding 43
nvidia/Nemotron-3-Nano-30B-A3B
262K ctx$0.05/1M
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B
262K ctx$0.10/1M
Coding 31
moonshotai/Kimi-K2.6
262K ctx$0.75/1M
Coding 47
gemma-4-31B-it
262K ctx$0.13/1M
deepseek-ai/DeepSeek-V4-Pro
1.0M ctx$1.30/1M
Coding 48

Frequently asked questions

How much does Qwen/Qwen2.5-72B-Instruct cost?
Qwen/Qwen2.5-72B-Instruct is priced at $0.23 per million input tokens and $0.40 per million output tokens when accessed via Requesty. Requesty charges exactly what the upstream provider charges β€” we don't add markup.
What is the context window of Qwen/Qwen2.5-72B-Instruct?
Qwen/Qwen2.5-72B-Instruct has a context window of 131K tokens. That's roughly 175 words of input you can fit in a single prompt.
How does Qwen/Qwen2.5-72B-Instruct perform on benchmarks?
Qwen/Qwen2.5-72B-Instruct scores 72.0% on MMLU Pro, 49.1% on GPQA Diamond, 34.5% on τ²-Bench. See the full benchmark chart above for results across MMLU Pro, GPQA Diamond, SWE-Bench Verified, HumanEval, MATH, AIME, MMMU, and LiveBench.
What can Qwen/Qwen2.5-72B-Instruct do?
Qwen/Qwen2.5-72B-Instruct supports tool calling, structured outputs (JSON schema). You can call it through any OpenAI-compatible client by pointing base_url to Requesty.
How do I use Qwen/Qwen2.5-72B-Instruct with the OpenAI SDK?
Install the OpenAI SDK, set base_url to "https://router.requesty.ai/v1", set your API key to your Requesty key, and set the model to "deepinfra/Qwen/Qwen2.5-72B-Instruct". The Quickstart above shows Python, JavaScript and cURL snippets.

Access Qwen/Qwen2.5-72B-Instruct through Requesty

One API key, 400+ models, OpenAI-compatible. No markup on provider prices, automatic failover, and smart caching built-in.

Get started freeAll DeepInfra Inc. models