Compare / Qwen3 8B vs Gemini 2.5 Pro
Qwen3 8B vs Gemini 2.5 Pro
Which is cheaper, which is more capable, and which does better at each kind of work, from 17 results on 3 sources that measured both models.
Qwen3 8B or Gemini 2.5 Pro? The short answer
- Qwen3 8B is cheaper: $0.20 against $3.44 per million tokens, 17 times less.
- Gemini 2.5 Pro scores higher on the Artificial Analysis Intelligence Index: 16.1 against 7.3.
- On our business benchmarks, Qwen3 8B ranks higher for sticking to the facts.
- On our business benchmarks, Gemini 2.5 Pro ranks higher for agents and tool use, reasoning and knowledge and writing.
At a glance
| Qwen3 8B | Gemini 2.5 Pro | |
|---|---|---|
| Price per million tokens | $0.12 in · $0.46 out | $1.25 in · $10.00 out |
| Blended (3 in : 1 out) | $0.20 | $3.44 |
| Cheapest host, blended | – | $1.72 (Google AI Studio) |
| Intelligence Index | 7.3 | 16.1 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | – | – |
| Context window | 131,072 tokens | 1,048,576 tokens |
| Weights | Open | Closed |
| Developer based in | China | the US |
Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.
Which is better for what
Each model's rank among models available today, for each kind of work, from our own benchmarks and licensed sources. The better rank is in green.
| Use case | Qwen3 8B | Gemini 2.5 Pro |
|---|---|---|
| Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls | 126 of 161 | 109 of 161 |
| Reasoning and knowledgeHard questions across science, maths and general knowledge | 166 of 195 | 92 of 195 |
| WritingWhat people prefer in blind comparisons, and judged writing quality | 158 of 187 | 42 of 187 |
| Sticking to the factsSummarising without inventing things, and factual answers | 7 of 153 | 38 of 153 |
Every shared result
| Benchmark | Metric | Qwen3 8B | Gemini 2.5 Pro |
|---|---|---|---|
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 9.00 | 33.3 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 7.30 | 16.1 |
| Artificial Analysisreported | GPQA Diamond% of questions | 58.9% | 84.4% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 3.9% | 22.5% |
| Artificial Analysisreported | IFBench% of instructions | 33.5% | 48.7% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 0.0% | 69.0% |
| Artificial Analysisreported | SciCode% of problems | 27.9% | 46.3% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 2.2% | 28.5% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 0.0% | 0.0% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 2.3% | 26.5% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 4.7% | 9.7% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 27.8% | 54.1% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 17.0% | 19.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.36 | 0.40 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 28.0 | 65.2 |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 99.9% | 99.1% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 4.8% | 7.0% |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.
Quick answers
Is Qwen3 8B better than Gemini 2.5 Pro?
It depends on the task. Among models available today, on our benchmarks and licensed sources, Qwen3 8B ranks higher for sticking to the facts; Gemini 2.5 Pro ranks higher for agents and tool use, reasoning and knowledge and writing.
Which is cheaper, Qwen3 8B or Gemini 2.5 Pro?
Qwen3 8B costs $0.20 per million tokens (three input to one output), against $3.44 for Gemini 2.5 Pro.
Run Qwen3 8B and Gemini 2.5 Pro on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.