Compare / Qwen3 235B A22B Instruct 2507 vs Qwen3 8B
Qwen3 235B A22B Instruct 2507 vs Qwen3 8B
Which is cheaper, which is more capable, and which does better at each kind of work, from 18 results on 3 sources that measured both models.
Qwen3 235B A22B Instruct 2507 or Qwen3 8B? The short answer
- Qwen3 8B is cheaper: $0.20 against $0.30 per million tokens, 33% less.
- Qwen3 235B A22B Instruct 2507 scores higher on the Artificial Analysis Intelligence Index: 12.0 against 7.3.
- On our business benchmarks, Qwen3 235B A22B Instruct 2507 ranks higher for agents and tool use, reasoning and knowledge and writing.
- On our business benchmarks, Qwen3 8B ranks higher for sticking to the facts.
At a glance
| Qwen3 235B A22B Instruct 2507 | Qwen3 8B | |
|---|---|---|
| Price per million tokens | $0.15 in · $0.75 out | $0.12 in · $0.46 out |
| Blended (3 in : 1 out) | $0.30 | $0.20 |
| Cheapest host, blended | $0.15 (GMICloud, FP8) | – |
| Intelligence Index | 12.0 | 7.3 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | – | – |
| Context window | 262,144 tokens | 131,072 tokens |
| Weights | Open | Open |
| Developer based in | China | China |
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 235B A22B Instruct 2507 | Qwen3 8B |
|---|---|---|
| Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls | 49 of 161 | 126 of 161 |
| Reasoning and knowledgeHard questions across science, maths and general knowledge | 125 of 195 | 166 of 195 |
| WritingWhat people prefer in blind comparisons, and judged writing quality | 83 of 187 | 158 of 187 |
| Sticking to the factsSummarising without inventing things, and factual answers | 97 of 153 | 7 of 153 |
Every shared result
| Benchmark | Metric | Qwen3 235B A22B Instruct 2507 | Qwen3 8B |
|---|---|---|---|
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 12.0 | 7.30 |
| Artificial Analysisreported | GPQA Diamond% of questions | 75.3% | 58.9% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 11.1% | 3.9% |
| Artificial Analysisreported | IFBench% of instructions | 46.1% | 33.5% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 33.9% | 0.0% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 15.2% | 2.3% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 33.3% | 27.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Irrelevance detection% correct | 81.7% | 82.3% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Memory% correct | 23.9% | 14.6% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Multi-turn tasks% correct | 45.4% | 41.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Overall accuracy% correct | 52.1% | 42.6% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Relevance detection% correct | 93.8% | 93.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (curated)% correct | 90.3% | 88.6% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (user-contributed)% correct | 78.7% | 80.5% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Web search% correct | 54.0% | 13.5% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 15.0% | 17.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.36 | 0.36 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 49.0 | 28.0 |
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 235B A22B Instruct 2507 better than Qwen3 8B?
It depends on the task. Among models available today, on our benchmarks and licensed sources, Qwen3 235B A22B Instruct 2507 ranks higher for agents and tool use, reasoning and knowledge and writing; Qwen3 8B ranks higher for sticking to the facts.
Which is cheaper, Qwen3 235B A22B Instruct 2507 or Qwen3 8B?
Qwen3 8B costs $0.20 per million tokens (three input to one output), against $0.30 for Qwen3 235B A22B Instruct 2507.
Run Qwen3 235B A22B Instruct 2507 and Qwen3 8B on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.