Compare / Qwen3-30B-A3B-Instruct-2507 vs Phi-4
Qwen3-30B-A3B-Instruct-2507 vs Phi-4
Which is cheaper, which is more capable, and which does better at each kind of work, from 24 results on 4 sources that measured both models.
Of the 24 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).
Qwen3-30B-A3B-Instruct-2507 or Phi-4? The short answer
They score about the same, so pick Phi-4: it costs 39% less per token.
- Phi-4 is cheaper: $0.088 against $0.14 per million tokens, 39% less.
- Qwen3-30B-A3B-Instruct-2507 scores higher on the Artificial Analysis Intelligence Index: 7.5 against 5.9.
- Qwen3-30B-A3B-Instruct-2507 is newer: released 29 Jul 2025, against 12 Dec 2024 for Phi-4.
- On third-party benchmarks, on the results both models have: Qwen3-30B-A3B-Instruct-2507 does better for agents and tool use, reasoning and knowledge and writing.
Head to head
Every fact side by side
| Qwen3-30B-A3B-Instruct-2507 | Phi-4 | |
|---|---|---|
| Price per million tokens | $0.090 in · $0.30 out | $0.070 in · $0.14 out |
| Blended (3 in : 1 out) | $0.14 | $0.088 |
| Cheapest host, blended | $0.084 (StreamLake) | – |
| Intelligence Index | 7.5 | 5.9 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | – | 44 |
| Context window | 262,144 tokens | 16,384 tokens |
| Released | 29 Jul 2025 (newer) | 12 Dec 2024 |
| Weights | Open | Open |
| 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 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 | Qwen3-30B-A3B-Instruct-2507 | Phi-4 |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results | ||
| Overall · TextArena rating, higher is better 162 · 291 of 387 measured | 1,383 | 1,256 |
| Business, management and finance · TextArena rating, higher is better 144 · 282 of 380 measured | 1,397 | 1,256 |
| Creative writing · TextArena rating, higher is better 186 · 290 of 385 measured | 1,319 | 1,210 |
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 209 · 231 of 243 measured | 7.5 | 5.9 |
| GPQA Diamond% of questions, higher is better 172 · 190 of 231 measured | 65.9% | 57.5% |
| Humanity's Last Exam% of questions, higher is better 173 · 221 of 242 measured | 6.9% | 3.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4 reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 3 results | ||
| Overall accuracy · BFCL% correct, higher is better 32 · 51 of 81 measured | 41.4%bfcl fc | 28.8%bfcl prompt |
| Irrelevance detection · BFCL% correct, higher is better 49 · 16 of 81 measured | 79.9%bfcl fc | 87.6%bfcl prompt |
| Memory · BFCL% correct, higher is better 38 · 24 of 81 measured | 17.6%bfcl fc | 24.7%bfcl prompt |
| UGI Leaderboard reported by UGI Leaderboard · 3 results | ||
| Writing score · UGIscore out of 100, higher is better 163 · 175 of 230 measured | 27.9 | 25.7 |
| Requested-length error · UGI% off the requested word count, lower is better 57 · 99 of 230 measured | 13.0% | 19.0% |
| Style adherence · UGIscore from 0 to 1, higher is better 187 · 122 of 230 measured | 0.31 | 0.34 |
Show all 24 results (12 more)
| Measure | Qwen3-30B-A3B-Instruct-2507 | Phi-4 |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results | ||
| Expert prompts · TextArena rating, higher is better 163 · 261 of 338 measured | 1,395 | 1,268 |
| Instruction following · TextArena rating, higher is better 164 · 285 of 387 measured | 1,367 | 1,245 |
| Writing, literature and language · TextArena rating, higher is better 176 · 290 of 386 measured | 1,342 | 1,229 |
| Artificial Analysis reported by Artificial Analysis · 4 results | ||
| IFBench% of instructions, higher is better 168 · 192 of 195 measured | 33.1% | 23.5% |
| Long-context reasoning (AA-LCR)% of questions, higher is better 191 · 222 of 233 measured | 26.3% | 0.0% |
| Terminal-Bench Hard% of tasks, higher is better 147 · 162 of 190 measured | 6.1% | 3.8% |
| τ²-bench telecom% of tasks, higher is better 185 · 188 of 193 measured | 10.2% | 0.0% |
| Berkeley Function Calling Leaderboard (BFCL) V4 reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 5 results | ||
| Multi-turn tasks · BFCL% correct, higher is better 34 · 64 of 81 measured | 30.0%bfcl fc | 3.9%bfcl prompt |
| Relevance detection · BFCL% correct, higher is better 8 · 74 of 81 measured | 93.8%bfcl prompt | 50.0%bfcl prompt |
| Single-turn calls (curated) · BFCL% correct, higher is better 9 · 69 of 81 measured | 88.9%bfcl prompt | 69.6%bfcl prompt |
| Single-turn calls (user-contributed) · BFCL% correct, higher is better 22 · 66 of 81 measured | 78.4%bfcl prompt | 60.7%bfcl prompt |
| Web search · BFCL% correct, higher is better 30 · 44 of 81 measured | 22.5%bfcl fc | 4.5%bfcl prompt |
Quick answers
Is Qwen3-30B-A3B-Instruct-2507 better than Phi-4?
They score about the same, so pick Phi-4: it costs 39% less per token.
Which is cheaper, Qwen3-30B-A3B-Instruct-2507 or Phi-4?
Phi-4 is cheaper: $0.088 against $0.14 per million tokens, 39% less.
Which is better for agents and tool use, Qwen3-30B-A3B-Instruct-2507 or Phi-4?
Qwen3-30B-A3B-Instruct-2507 ranks 75 of 164 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 123 for Phi-4.
Which is better for reasoning and knowledge, Qwen3-30B-A3B-Instruct-2507 or Phi-4?
Qwen3-30B-A3B-Instruct-2507 ranks 159 of 198 models on sale for hard questions across science, maths and general knowledge, against 181 for Phi-4.
Which is better for writing, Qwen3-30B-A3B-Instruct-2507 or Phi-4?
Qwen3-30B-A3B-Instruct-2507 ranks 140 of 188 models on sale for what people prefer in blind comparisons, and judged writing quality, against 176 for Phi-4.
Which has the bigger context window, Qwen3-30B-A3B-Instruct-2507 or Phi-4?
Qwen3-30B-A3B-Instruct-2507: 262,144 tokens, against 16,384 for Phi-4.
Run Qwen3-30B-A3B-Instruct-2507 and Phi-4 on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.