Compare / Qwen3-235B-A22B-Instruct-2507 vs Qwen3-4B-Instruct-2507
Qwen3-235B-A22B-Instruct-2507 vs Qwen3-4B-Instruct-2507
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.
Of the 18 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.
Results as of 9 October 2026 (catalogue release 2026-10-09-dd620e778ac5).
Qwen3-235B-A22B-Instruct-2507 or Qwen3-4B-Instruct-2507? The short answer
Pick Qwen3-235B-A22B-Instruct-2507: it scores higher (Intelligence Index 12.0 against 6.7) at about the same price.
- Qwen3-235B-A22B-Instruct-2507 scores higher on the Artificial Analysis Intelligence Index: 12.0 against 6.7.
- Qwen3-4B-Instruct-2507 is newer: released 6 Aug 2025, against 21 Jul 2025 for Qwen3-235B-A22B-Instruct-2507.
- On third-party benchmarks, on the results both models have: Qwen3-235B-A22B-Instruct-2507 does better for agents and tool use, reasoning and knowledge and writing.
Head to head
Every fact side by side
| Qwen3-235B-A22B-Instruct-2507 | Qwen3-4B-Instruct-2507 | |
|---|---|---|
| Price per million tokens | $0.15 in · $0.75 out | Not on sale |
| Blended (3 in : 1 out) | $0.30 | – |
| Cheapest host, blended | $0.15 (GMICloud, FP8) | – |
| Intelligence Index | 12.0 | 6.7 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | – | – |
| Context window | 262,144 tokens | – |
| Released | 21 Jul 2025 | 6 Aug 2025 (newer) |
| Weights | Open | Closed |
| 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 the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.
Every shared result
Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.
| Measure | Qwen3-235B-A22B-Instruct-2507 | Qwen3-4B-Instruct-2507 |
|---|---|---|
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 153 · 222 of 243 measured | 12.0 | 6.7 |
| GPQA Diamond% of questions, higher is better 136 · 198 of 231 measured | 75.3% | 51.7% |
| Humanity's Last Exam% of questions, higher is better 144 · 202 of 242 measured | 11.1% | 4.5% |
| Berkeley Function Calling Leaderboard (BFCL) V4 reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 3 results | ||
| Overall accuracy · BFCL% correct, higher is better 20 · 40 of 81 measured | 52.2%bfcl prompt | 35.7%bfcl fc |
| Irrelevance detection · BFCL% correct, higher is better 41 · 30 of 81 measured | 81.7%bfcl fc | 84.9%bfcl fc |
| Memory · BFCL% correct, higher is better 26 · 26 of 81 measured | 23.9%bfcl fc | 23.9%bfcl prompt |
| UGI Leaderboard reported by UGI Leaderboard · 3 results | ||
| Writing score · UGIscore out of 100, higher is better 80 · 154 of 230 measured | 49.0 | 29.9 |
| Requested-length error · UGI% off the requested word count, lower is better 69 · 25 of 230 measured | 15.0% | 8.0% |
| Style adherence · UGIscore from 0 to 1, higher is better 64 · 117 of 230 measured | 0.36 | 0.35 |
Show all 18 results (9 more)
| Measure | Qwen3-235B-A22B-Instruct-2507 | Qwen3-4B-Instruct-2507 |
|---|---|---|
| Artificial Analysis reported by Artificial Analysis · 4 results | ||
| IFBench% of instructions, higher is better 113 · 164 of 195 measured | 46.1% | 33.5% |
| Long-context reasoning (AA-LCR)% of questions, higher is better 177 · 214 of 233 measured | 33.9% | 11.3% |
| Terminal-Bench Hard% of tasks, higher is better 110 · 156 of 190 measured | 15.2% | 4.5% |
| τ²-bench telecom% of tasks, higher is better 127 · 146 of 193 measured | 33.3% | 26.6% |
| Berkeley Function Calling Leaderboard (BFCL) V4 reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 5 results | ||
| Multi-turn tasks · BFCL% correct, higher is better 20 · 40 of 81 measured | 45.4%bfcl fc | 22.1%bfcl fc |
| Relevance detection · BFCL% correct, higher is better 8 · 26 of 81 measured | 93.8%bfcl prompt | 87.5%bfcl fc |
| Single-turn calls (curated) · BFCL% correct, higher is better 2 · 21 of 81 measured | 90.3%bfcl prompt | 87.9%bfcl fc |
| Single-turn calls (user-contributed) · BFCL% correct, higher is better 15 · 28 of 81 measured | 78.7%bfcl prompt | 76.4%bfcl fc |
| Web search · BFCL% correct, higher is better 20 · 44 of 81 measured | 54.0%bfcl fc | 4.5%bfcl prompt |
Quick answers
Is Qwen3-235B-A22B-Instruct-2507 better than Qwen3-4B-Instruct-2507?
Pick Qwen3-235B-A22B-Instruct-2507: it scores higher (Intelligence Index 12.0 against 6.7) at about the same price.
Which is better for agents and tool use, Qwen3-235B-A22B-Instruct-2507 or Qwen3-4B-Instruct-2507?
Qwen3-235B-A22B-Instruct-2507 ranks 51 of 164 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 97 for Qwen3-4B-Instruct-2507.
Which is better for reasoning and knowledge, Qwen3-235B-A22B-Instruct-2507 or Qwen3-4B-Instruct-2507?
Qwen3-235B-A22B-Instruct-2507 ranks 129 of 198 models on sale for hard questions across science, maths and general knowledge, against 172 for Qwen3-4B-Instruct-2507.
Which is better for writing, Qwen3-235B-A22B-Instruct-2507 or Qwen3-4B-Instruct-2507?
Qwen3-235B-A22B-Instruct-2507 ranks 85 of 188 models on sale for what people prefer in blind comparisons, and judged writing quality, against 149 for Qwen3-4B-Instruct-2507.
Run Qwen3-235B-A22B-Instruct-2507 and Qwen3-4B-Instruct-2507 on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.