Qwen2.5-Coder-3B-Instruct
An instruction-tuned 3.09B-parameter model from the Qwen2.5-Coder series, the latest series of code-specific Qwen large language models.
text + code → text + code · made by Alibaba
$0.01→$0.03per Mtok in / out
Nscale · 3 sellers
Sold by 3 ways
| Seller | Lane | Rate |
|---|---|---|
| Hugging Face Inference Providers aggregator | nscale | $0.01per Mtok in$0.03per Mtok outrouter.huggingface.co · read 2026-08-25 |
| Nscale aggregator | standard | $0.01per Mtok in$0.03per Mtok outraw.githubusercontent.com · read 2026-09-16 |
| Fireworks AI aggregator | standard | $0.1per Mtok in$0.1per Mtok outraw.githubusercontent.com · read 2026-09-16 |
Measured 1 standing
| Place | Board | Metric | Score |
|---|---|---|---|
| 21stof 170 | Gsm8k — Epoch AI | EM | 0.807 |
About
Qwen2.5-Coder-3B-Instruct — a text and code model from Alibaba, sold by 3 companies from $0.01 in and $0.03 out per million tokens, placed 21st of 170 on Gsm8k — Epoch AI.
It takes text and code and returns text and code, with a context window of 32,768 tokens. The catalogue files it under code. Three companies sell it. The cheapest is $0.01 in and $0.03 out per million tokens at Nscale. Beside the standard rate there is a separately routed lane. It stands 21st of 170 on Gsm8k — Epoch AI.
Every current figure
- Maker
- Alibaba
- Register
- model
- Takes
- text + code
- Returns
- text + code
- Context
- 32,768 tokens
- Parameters
- 3 billion · read from its own name
- Licence
- not read
- Sellers
- 3
- Maker's own price
- not read
- Price
- $0.01 in and $0.03 out per million tokens — Nscale
- Boards
- 1
- Best place
- 21st of 170 — Gsm8k — Epoch AI
Known as 1 name
Qwen/Qwen2.5-Coder-3B-Instruct