Phi 4 vs Qwen3.5-122B-A10B
Qwen3.5-122B-A10B leads Phi 4 on the Artificial Analysis Intelligence Index, 17.7 to 5.9, and Phi 4 costs 14.9x less per output token ($0.14 vs $2.08 per million). Updated October 21, 2026; 34 benchmarks have a result for both.
Phi 4 or Qwen3.5-122B-A10B: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick Phi 4 for output tokens 14.9x cheaper ($0.14 vs $2.08 per million) and work measured by Terminal-Bench Hard (Artificial Analysis run) or LMArena Text: coding, where it leads.
- Shared results
- 34
- Phi 4 only
- 12
- Qwen3.5-122B-A10B only
- 17
- Like-for-like categories
- 7 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
No call
Neither has a coding score yet.
One side not measured
Intelligence
AA Intelligence Index
Qwen3.5-122B-A10B
Qwen3.5-122B-A10B leads, 17.7 against 5.9.
Same board, both measured
Long documents
Listed context window
Qwen3.5-122B-A10B
Qwen3.5-122B-A10B takes 262K tokens against 16K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Phi 4
Phi 4 $0.0001, Qwen3.5-122B-A10B $0.0013 per turn.
List prices
At a glance
- Coding index
- Phi 4 not benchmarked, Qwen3.5-122B-A10B not benchmarked.
- Widest category
- Agentic work, 93.6 points apart on average over 1 shared benchmark.
- Widest single gap
- τ²-Bench Telecom: Phi 4 0.0% against Qwen3.5-122B-A10B 93.6%, 93.6 points apart on a 0 to 100 scale.
- Output price
- Phi 4 $0.14 against Qwen3.5-122B-A10B $2.08 per million output tokens, 14.9x apart.
- Context window
- Phi 4 takes 16K tokens, Qwen3.5-122B-A10B 262K.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | Phi 4 | Bars | Qwen3.5-122B-A10B | Ties |
|---|---|---|---|---|
| Coding | 0 | 3 | ||
| Intelligence | 0 | 1 | ||
| Agentic work | 0 | 1 | ||
| Reasoning | 0 | 5 | ||
| Math | 0 | 2 | ||
| Knowledge | 0 | 4 | ||
| General | 0 | 18 | ||
| All shared boards | 0 | 34 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index03
| Benchmark | Phi 4 | Qwen3.5-122B-A10B | Leader |
|---|---|---|---|
| LMArena WebDev | Not measured | 1361 | Only Qwen3.5-122B-A10B measured |
| AA Coding Index | Not measured | 43.3 | Only Qwen3.5-122B-A10B measured |
| LiveCodeBench (Artificial Analysis run) | 23.1% | Not measured | Only Phi 4 measured |
| SciCode | Not measured | 39.7% | Only Qwen3.5-122B-A10B measured |
| Terminal-Bench Hard (Artificial Analysis run) | 3.8% | 31.1% | Qwen3.5-122B-A10B |
| LMArena Text: coding | 1231 | 1436 | Qwen3.5-122B-A10B |
| LMArena Text: software and IT | 1228 | 1436 | Qwen3.5-122B-A10B |
| LMArena WebDev: React | Not measured | 1353 | Only Qwen3.5-122B-A10B measured |
| LMArena WebDev: HTML | Not measured | 1351 | Only Qwen3.5-122B-A10B measured |
| Neither model has a result on the other 30 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Phi 4 | Qwen3.5-122B-A10B | Leader |
|---|---|---|---|
| AA Intelligence Index | 5.9 | 17.7 | Qwen3.5-122B-A10B |
| Epoch Capabilities Index | 130.5 | Not measured | Only Phi 4 measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Phi 4 | Qwen3.5-122B-A10B | Leader |
|---|---|---|---|
| τ²-Bench Telecom | 0.0% | 93.6% | Qwen3.5-122B-A10B |
| BALROG | 11.6% | Not measured | Only Phi 4 measured |
| Berkeley Function Calling Leaderboard (BFCL) | 28.8% | Not measured | Only Phi 4 measured |
| Neither model has a result on the other 38 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index05
| Benchmark | Phi 4 | Qwen3.5-122B-A10B | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 57.5% | 85.7% | Qwen3.5-122B-A10B |
| Humanity's Last Exam (Artificial Analysis run) | 3.8% | 25.2% | Qwen3.5-122B-A10B |
| GPQA Diamond (Epoch AI run) | 56.1% | Not measured | Only Phi 4 measured |
| Mystery game puzzles | Not measured | 17.0% | Only Qwen3.5-122B-A10B measured |
| LMArena Text: hard prompts | 1220 | 1421 | Qwen3.5-122B-A10B |
| LMArena Text: expert | 1203 | 1431 | Qwen3.5-122B-A10B |
| LMArena Text: hard prompts in English | 1230 | 1431 | Qwen3.5-122B-A10B |
| LMArena Vision: diagrams | Not measured | 1249 | Only Qwen3.5-122B-A10B measured |
| LMArena Vision: homework | Not measured | 1261 | Only Qwen3.5-122B-A10B measured |
| Neither model has a result on the other 13 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Phi 4 | Qwen3.5-122B-A10B | Leader |
|---|---|---|---|
| AA Math Index | 18.0 | Not measured | Only Phi 4 measured |
| AIME 2025 (Artificial Analysis run) | 18.0% | Not measured | Only Phi 4 measured |
| OTIS Mock AIME | 13.8% | Not measured | Only Phi 4 measured |
| MATH level 5 | 64.9% | Not measured | Only Phi 4 measured |
| LMArena Text: math | 1246 | 1431 | Qwen3.5-122B-A10B |
| LMArena Text: mathematical work | 1254 | 1431 | Qwen3.5-122B-A10B |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index04
| Benchmark | Phi 4 | Qwen3.5-122B-A10B | Leader |
|---|---|---|---|
| MMLU Pro (Artificial Analysis run) | 71.4% | Not measured | Only Phi 4 measured |
| MMLU | 84.8% | Not measured | Only Phi 4 measured |
| LMArena Text: science | 1206 | 1432 | Qwen3.5-122B-A10B |
| LMArena Text: medicine and healthcare | 1202 | 1423 | Qwen3.5-122B-A10B |
| LMArena Text: legal and government | 1241 | 1422 | Qwen3.5-122B-A10B |
| LMArena Text: business and finance | 1199 | 1414 | Qwen3.5-122B-A10B |
| Neither model has a result on the other 12 boards here | |||
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Phi 4 | Qwen3.5-122B-A10B | Leader |
|---|---|---|---|
| Neither model has a result on 9 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index018
| Benchmark | Phi 4 | Qwen3.5-122B-A10B | Leader |
|---|---|---|---|
| IFBench | 23.5% | 75.7% | Qwen3.5-122B-A10B |
| Long context reasoning (AA-LCR) | 0.0% | 76.3% | Qwen3.5-122B-A10B |
| DTBench | Not measured | 84.3% | Only Qwen3.5-122B-A10B measured |
| LMCA | Not measured | 32.2% | Only Qwen3.5-122B-A10B measured |
| Creative writing (Lech Mazur) | 62.6% | Not measured | Only Phi 4 measured |
| LMArena Text | 1217 | 1417 | Qwen3.5-122B-A10B |
| LMArena Text: instruction following | 1201 | 1399 | Qwen3.5-122B-A10B |
| LMArena Text: creative writing | 1181 | 1368 | Qwen3.5-122B-A10B |
| LMArena Vision | Not measured | 1244 | Only Qwen3.5-122B-A10B measured |
| LMArena Text: English | 1231 | 1428 | Qwen3.5-122B-A10B |
| LMArena Text: non-English | 1197 | 1401 | Qwen3.5-122B-A10B |
| LMArena Text: Chinese | 1212 | 1463 | Qwen3.5-122B-A10B |
| LMArena Text: German | 1222 | 1425 | Qwen3.5-122B-A10B |
| LMArena Text: Japanese | 1158 | 1367 | Qwen3.5-122B-A10B |
| LMArena Text: Korean | 1151 | 1353 | Qwen3.5-122B-A10B |
| LMArena Text: French | 1224 | 1443 | Qwen3.5-122B-A10B |
| LMArena Text: Spanish | 1234 | 1425 | Qwen3.5-122B-A10B |
| LMArena Text: Russian | 1209 | 1400 | Qwen3.5-122B-A10B |
| LMArena Text: Polish | Not measured | 1398 | Only Qwen3.5-122B-A10B measured |
| LMArena Text: multi-turn | 1206 | 1416 | Qwen3.5-122B-A10B |
| LMArena Text: longer queries | 1217 | 1410 | Qwen3.5-122B-A10B |
| LMArena Text: writing and language | 1198 | 1387 | Qwen3.5-122B-A10B |
| LMArena Text: entertainment and media | 1162 | 1366 | Qwen3.5-122B-A10B |
| LMArena Vision: English | Not measured | 1249 | Only Qwen3.5-122B-A10B measured |
| LMArena Vision: Chinese | Not measured | 1301 | Only Qwen3.5-122B-A10B measured |
| LMArena Vision: reading text | Not measured | 1251 | Only Qwen3.5-122B-A10B measured |
| LMArena Vision: humor | Not measured | 1243 | Only Qwen3.5-122B-A10B measured |
| LMArena Vision: creative writing | Not measured | 1220 | Only Qwen3.5-122B-A10B measured |
| Neither model has a result on the other 18 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Phi 4 | Qwen3.5-122B-A10B | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
Phi 4 costs less
Repository review
50K input + 3K output tokens
Phi 4 costs less
Agent loop
200K input + 10K output tokens
Phi 4 costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | Microsoft Phi 4 | Qwen3.5-122B-A10B |
|---|---|---|
| Maker | Microsoft | Qwen |
| Weights | Open weights | Open weights |
| Released | January 10, 2025 | February 25, 2026 |
| Context window | 16K | 262K |
| Input $/Mtok | $0.07 | $0.26 |
| Output $/Mtok | $0.14 | $2.08 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 0 | 1 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Phi 4 better than Qwen3.5-122B-A10B for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Phi 4 for output tokens 14.9x cheaper ($0.14 vs $2.08 per million) and work measured by Terminal-Bench Hard (Artificial Analysis run) or LMArena Text: coding, where it leads.
Which is cheaper, Phi 4 or Qwen3.5-122B-A10B?
Phi 4 costs $0.07 per million input tokens and $0.14 per million output tokens; Qwen3.5-122B-A10B costs $0.26 and $2.08. Cost per solved task is known for neither of them.
Which has the bigger context window, Phi 4 or Qwen3.5-122B-A10B?
Qwen3.5-122B-A10B, with 262K tokens against 16K.
How many benchmarks is the Phi 4 vs Qwen3.5-122B-A10B comparison based on?
34 benchmarks have a result for both models as of October 21, 2026. Phi 4's coding index rests on 0 independent benchmarks and Qwen3.5-122B-A10B's on 1. Where only one model has a result, the page says so instead of naming a winner.