Qwen2.5 7B Instruct
Alibaba · Released Oct 2024
Qwen2.5 7B Instruct is a 7-billion-parameter instruction-tuned dense language model from Alibaba with a 32,768-token context window.
- Strengths
- The model runs efficiently in resource-constrained environments while retaining reasonable instruction-following capability for its size.
- Best for
- Latency-sensitive applications and on-device or edge deployments where parameter count matters more than absolute performance.
- Limitations
- Substantially smaller than the Qwen3 lineup (released in 2025) and earlier Qwen2.5 dense models like the 72B variant, making it less suitable for complex reasoning, code generation, or nuanced language understanding tasks.
Input / 1M
$0.1
Output / 1M
$0.2
Cached input / 1M
—
Context window
32K
Price history
Input (solid)Output (dashed)
Price change
- 30d
- in — out —
- 90d
- in increased 150.0% out increased 100.0%
- 1y
- in increased 150.0% out increased 100.0%
- Since launch
- in increased 150.0% out increased 100.0%
Snapshots
| Effective | Input | Output | Cached in | Note | Source |
|---|---|---|---|---|---|
| 30 Jul 2026 | $0.1 | $0.2 | — | Imported from OpenRouter | openrouter.ai |
| 11 Jun 2026 | $0.04 | $0.1 | — | Imported from OpenRouter | openrouter.ai |