LANGUAGE MODEL Alibaba Last updated:

Qwen-7B / Qwen-14B

Alibaba's First Major Open Model

Alibaba's first publicly-released open-weight foundation model family — Qwen-7B and Qwen-14B — released under a permissive commercial license. Outperformed Llama 2 on both Chinese (by a wide margin) and English benchmarks at comparable size. Established Alibaba as a credible open-LLM player and seeded the Qwen brand that would dominate Chinese open-source LLMs through 2026.

Cost
Free
Open weights — self-host
Context
8K
Up to 8,192 tokens
How are Intelligence, Speed & Cost bucketed?
Intelligence and Speed buckets are percentile ranks on Artificial Analysis. Cost buckets are fixed dollar thresholds keyed off output-token price ($/M out).
Intelligence
  • Top 1%≤ 1%
  • Top 5%≤ 5%
  • Top 10%≤ 10%
  • Good≤ 25%
  • Medium≤ 50%
  • Below avg> 50%
Speed
  • Top 1%≥ 345 tok/s
  • Top 5%≥ 237 tok/s
  • Top 10%≥ 196 tok/s
  • Good≥ 146 tok/s
  • Medium≥ 90 tok/s
  • Slow< 90 tok/s
Cost
  • Freeopen weights · self-host
  • Low< $1 / M out
  • Moderate$1–5 / M out
  • High≥ $5 / M out

Why it matters

Qwen normalized the expectation that Chinese hyperscalers would publish frontier-comparable models openly — a pattern continued by DeepSeek, ByteDance Doubao (partial), and Tencent Hunyuan. The pattern is structurally different from US hyperscalers (where Meta is the only major open-weight publisher).

Core Capabilities

Long Documents
Handles entire codebases, books, and multi-doc RAG.
Generative
Produces images, video, audio, or other media.

Context Window

8k tokens
≈ short doc
4k Chat 聊天
32k Long docs 长文档
128k Books 整本书
400k Multi-doc 多文档
1M Codebase 整个代码库
10M
8k

Availability

API
Not available
Product / App
Not available
Open Source
Released
Enterprise

Pricing Model

Free / self-host
Open weights — pay only for compute.
Self-host

Capability / Performance

Where this model sits relative to the middle 60% of models in the tree. All scores are 0–10 (higher is better).

Lower 20% Upper 80% This model
Reasoning
AA Intelligence Index · scaled to 10
1.7
5.6
1.4
Context / memory
Context window size · log-scaled
6.0
9.0
2.0
Lower 20% 20th percentile — 20% of models score below this This model Where the current model lands Upper 80% 80th percentile — only 20% of models score above this Percentile boundaries are computed across every model in the tree that reports the underlying benchmark for each capability.

What it feels like

  • First-generation Qwen — historical; superseded by 2.5+.
  • First open-source model to top OpenCompass leaderboard — beat Llama 3.1 405B with 1/5 the parameters
  • 72B-Instruct: 74.2 on coding, 77 on math — outscored Claude 3.5 Sonnet (72.1) and GPT-4o (70.6) at release
  • MMLU 85+, HumanEval 85+, MATH 80+ — frontier-tier across the board for an open-weights model

Best use cases

  • Self-hosted production where Llama 3.1 405B is too big to serve
  • Multilingual deployments (29 languages with strong coverage)
  • Fine-tuning on private code or technical corpora

Tools to try

Not ideal for

  • Frontier reasoning by mid-2025 — Qwen 3 / DeepSeek R1 / Claude 4 series have moved past it
  • Vision tasks (use Qwen 2.5-VL) or audio (use Qwen 2.5-Audio)

Model Evolution

View full evolution tree →