Qwen3 8B
Qwen3-8B is a dense 8.2B parameter causal language model from the Qwen3 series, designed for both reasoning-heavy tasks and efficient dialogue.
text → text · made by Alibaba
Sold by 10 ways
| Seller | Lane | Rate |
|---|---|---|
| Thinking Machines Lab api | training | $0.2per Mtok in$0.6per Mtok out$0.039per Mtok cachedtinker-docs.thinkingmachines.ai · read 2026-08-25 |
| Novita AI 0 ways | $0.035per Mtok in$0.14per Mtok out | |
| Novita AI aggregatornot on the seller's list today | standard | $0.035per Mtok in$0.14per Mtok outapi.novita.ai · read 2026-08-25 |
| SiliconFlow aggregator | standard | $0.06per Mtok in$0.06per Mtok outmodels.dev · read 2026-10-31 |
| Hugging Face Inference Providers 2 ways | $0.07per Mtok in$0.18per Mtok out | |
| Hugging Face Inference Providers aggregator | standard | $0.07per Mtok in$0.18per Mtok outmodels.dev · read 2026-10-31 |
| Hugging Face Inference Providers aggregator | nscale | $0.07per Mtok in$0.18per Mtok outrouter.huggingface.co · read 2026-08-25 |
| Featherless AI aggregator | standard | $0.12per Mtok in$0.46per Mtok outapi.featherless.ai · read 2026-09-20 |
| Inference.net 0 ways | $0.12per Mtok in$0.46per Mtok out | |
| Inference.net aggregatornot on the seller's list today | standard | $0.12per Mtok in$0.46per Mtok outapi.inference.net · read 2026-10-18 |
| Nous Research 0 ways | $0.12per Mtok in$0.46per Mtok out | |
| Nous Research aggregatornot on the seller's list today | standard | $0.12per Mtok in$0.46per Mtok outinference-api.nousresearch.com · read 2026-09-12 |
| OpenRouter 0 ways | $0.12per Mtok in$0.46per Mtok out | |
| OpenRouter aggregatornot on the seller's list today | standard | $0.12per Mtok in$0.46per Mtok outopenrouter.ai · read 2026-08-24 |
| Fireworks AI aggregator | standard | $0.2per Mtok in$0.2per Mtok outraw.githubusercontent.com · read 2026-10-31 |
Measured 7 standings
| Place | Board | Metric | Score |
|---|---|---|---|
| 10thof 111 | Vectara Hallucination Leaderboard | Hallucination Rate | 4.8 |
| 184thof 306 | OTIS Mock AIME 2024-2025 | mean_score | 56.111 |
| 188thof 197 | Lmca — Epoch AI | Score | 8.82 |
| 194thof 235 | Dtbench — Epoch AI | Accuracy | 59.73 |
| 223rdof 326 | GPQA Diamond | mean_score | 56.755 |
| 230thof 235 | Scicode — Epoch AI | Score | 22.569 |
| 240thof 260 | Critpt — Epoch AI | Accuracy | 0 |
About
Qwen3 8B — a text model from Alibaba, sold by 5 companies from $0.06 in and $0.06 out per million tokens, placed 10th of 111 on Vectara Hallucination Leaderboard.
It takes text and returns text, with a context window of 32,768 tokens. It was published in April 2025, trained on material up to April 2025. Its sellers say it can reason step by step and call a tool. The catalogue files it under reasoning. Five companies sell it. The cheapest is $0.06 in and $0.06 out per million tokens at SiliconFlow. Beside the standard rate there is a separately routed lane. It has been measured on 7 boards, and stands best at 10th of 111 on Vectara Hallucination Leaderboard.
Every current figure
- Maker
- Alibaba
- Register
- model
- Takes
- text
- Returns
- text
- Context
- 32,768 tokens
- Published
- April 2025
- Knowledge to
- April 2025
- Parameters
- 8.2 billion
- Licence
- not read
- Sellers
- 5
- Maker's own price
- not read
- Price
- $0.06 in and $0.06 out per million tokens — SiliconFlow
- Boards
- 7
- Best place
- 10th of 111 — Vectara Hallucination Leaderboard
Known as 5 names
PrimeIntellect/Qwen3-8BQwen/Qwen3-8Bunsloth/Qwen3-8Bwillcb/Qwen3-8Bqwen/qwen3-8b-fp8