# Qwen2.5-1M

Qwen2.5-1M is an AI model released by Qwen on Jan 27 2025. It has 7B/14B parameters, a 1M context window and open weight. Tracked results include 38.6% on LiveCodeBench.

## Facts

| Field | Value |
| --- | --- |
| Model | Qwen2.5-1M |
| Developer | Qwen |
| Release date | Monday, Jan 27 2025 |
| Licensing | Open Weight |
| Parameters | 7B/14B |
| Context window | 1M |

## Tracked benchmark scores

| Benchmark | Score | Source | What it measures |
| --- | --- | --- | --- |
| LiveCodeBench | 38.6% | Lab | Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better. |
| HumanEval | 88.4% | Lab | 164 small Python problems: the AI is given a function's description and has to write the working function. This was the coding benchmark of the GPT-3.5 and GPT-4 era, before the field moved to fixing real bugs in real repositories. Higher is better. |
| IFEval | 84.3% | Lab | following verifiable instructions in generated responses |
| MATH | 79.5% | Lab | accuracy on the original competition mathematics test set |
| MMLU-Pro | 63.3% | Lab | accuracy on the ten-option MMLU-Pro benchmark |
| GSM8K | 94.8% | Lab | Grade-school maths word problems that take a few steps of arithmetic to work through. It separated the models of 2022 and 2023 sharply, then saturated. One caveat on the historical numbers: OpenAI included part of the GSM8K training set in GPT-4's pre-training mix, so GPT-4's score is not a clean few-shot result. Higher is better. |

## About Qwen2.5-1M

Qwen2.5-1M shipped on January 27, 2025 with two instruction-tuned checkpoints, in 7B and 14B sizes. It brought the million-token context that Alibaba had introduced on the Turbo API to downloadable Qwen weights.

Qwen published a vLLM-based inference implementation using sparse attention to reduce the cost of processing long inputs. Its short-context evaluation recorded 88.4 on HumanEval for the 14B checkpoint, showing how the long-context tuning affected a familiar coding task.

## Questions and answers

### When was Qwen2.5-1M released?

Qwen2.5-1M was released by Qwen on Monday, Jan 27 2025.

### Who made Qwen2.5-1M?

Qwen2.5-1M was built by Qwen. Alibaba's AI lab, building the Qwen family. The most prolific publisher of open-weight models of any major lab, alongside a closed Max and Plus tier sold through Alibaba Cloud.

### What benchmark scores did Qwen2.5-1M get?

Qwen2.5-1M reports 6 tracked benchmark scores — LiveCodeBench: 38.6%; HumanEval: 88.4%; IFEval: 84.3%; MATH: 79.5%; MMLU-Pro: 63.3%; GSM8K: 94.8%. Tracked scores may come from lab reports or independent benchmarks; source details accompany the benchmark data.

### What is the context window of Qwen2.5-1M?

Qwen2.5-1M has a context window of 1M. Providers may specify separate input and output token limits.

### How many parameters does Qwen2.5-1M have?

Qwen2.5-1M is reported at 7B/14B parameters.

### Is Qwen2.5-1M open source?

Partly. Qwen2.5-1M is an open-weight model: the trained weights are free to download and run locally or on your own infrastructure, but the training data and code are not fully released and the license may restrict some commercial uses. It is not open source in the strict sense.

### What came before and after Qwen2.5-1M?

Qwen's previous tracked release was QwQ-32B-Preview on Nov 28 2024, 60 days earlier. It was followed by Qwen2.5-Max on Jan 28 2025.


---

Canonical page: https://aireleasetracker.com/model/qwen/qwen2.5-1m
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.
