Qwen3 235B A22B Instruct 2507 vs Step 3.7 Flash
Step 3.7 Flash leads Qwen3 235B A22B Instruct 2507 on the Artificial Analysis Intelligence Index, 19.5 to 12.0, and Qwen3 235B A22B Instruct 2507 costs 1.5x less per output token ($0.75 vs $1.15 per million). Updated October 22, 2026; 14 benchmarks have a result for both.
Qwen3 235B A22B Instruct 2507 or Step 3.7 Flash: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick Qwen3 235B A22B Instruct 2507 for output tokens 1.5x cheaper ($0.75 vs $1.15 per million) and work measured by SciCode or Terminal-Bench Hard (Artificial Analysis run), where it leads.
- Shared results
- 14
- Qwen3 235B A22B Instruct 2507 only
- 10
- Step 3.7 Flash only
- 4
- Like-for-like categories
- 6 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
Step 3.7 Flash
Step 3.7 Flash leads, 19.5 against 12.0.
Same board, both measured
Web design
Design Arena: websites
Step 3.7 Flash
Step 3.7 Flash leads, 1196 against 1058.
Same board, both measured
Long documents
Listed context window
No call
Both take 262K tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen3 235B A22B Instruct 2507
Qwen3 235B A22B Instruct 2507 $0.0005, Step 3.7 Flash $0.0008 per turn.
List prices
At a glance
- Coding index
- Qwen3 235B A22B Instruct 2507 not benchmarked, Step 3.7 Flash not benchmarked.
- Widest category
- Agentic work, 45.3 points apart on average over 1 shared benchmark.
- Widest single gap
- τ²-Bench Telecom: Qwen3 235B A22B Instruct 2507 53.2% against Step 3.7 Flash 98.5%, 45.3 points apart on a 0 to 100 scale.
- Output price
- Qwen3 235B A22B Instruct 2507 $0.75 against Step 3.7 Flash $1.15 per million output tokens, 1.5x apart.
- Context window
- Both take 262K tokens.
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 | Qwen3 235B A22B Instruct 2507 | Bars | Step 3.7 Flash | Ties |
|---|---|---|---|---|
| Coding | 0 | 2 | ||
| Intelligence | 0 | 1 | ||
| Agentic work | 0 | 1 | ||
| Reasoning | 0 | 2 | ||
| Web design | 0 | 6 | ||
| General | 0 | 2 | ||
| All shared boards | 0 | 14 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index02
| Benchmark | Qwen3 235B A22B Instruct 2507 | Step 3.7 Flash | Leader |
|---|---|---|---|
| AA Coding Index | Not measured | 39.6 | Only Step 3.7 Flash measured |
| LiveCodeBench (Artificial Analysis run) | 78.8% | Not measured | Only Qwen3 235B A22B Instruct 2507 measured |
| SciCode | 41.4% | 43.9% | Step 3.7 Flash |
| Terminal-Bench Hard (Artificial Analysis run) | 15.2% | 35.6% | Step 3.7 Flash |
| WeirdML | 38.7% | Not measured | Only Qwen3 235B A22B Instruct 2507 measured |
| ALE-Bench | Not measured | 694.1 | Only Step 3.7 Flash measured |
| Neither model has a result on the other 33 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Qwen3 235B A22B Instruct 2507 | Step 3.7 Flash | Leader |
|---|---|---|---|
| AA Intelligence Index | 12.0 | 19.5 | Step 3.7 Flash |
| Neither model has a result on the other 1 board here | |||
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Qwen3 235B A22B Instruct 2507 | Step 3.7 Flash | Leader |
|---|---|---|---|
| τ²-Bench Telecom | 53.2% | 98.5% | Step 3.7 Flash |
| Berkeley Function Calling Leaderboard (BFCL) | 52.1% | Not measured | Only Qwen3 235B A22B Instruct 2507 measured |
| Neither model has a result on the other 39 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Qwen3 235B A22B Instruct 2507 | Step 3.7 Flash | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 79.0% | 80.9% | Step 3.7 Flash |
| Humanity's Last Exam (Artificial Analysis run) | 15.9% | 21.4% | Step 3.7 Flash |
| ARC-AGI | 11.0% | Not measured | Only Qwen3 235B A22B Instruct 2507 measured |
| ARC-AGI-2 | 1.3% | Not measured | Only Qwen3 235B A22B Instruct 2507 measured |
| Neither model has a result on the other 18 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | Qwen3 235B A22B Instruct 2507 | Step 3.7 Flash | Leader |
|---|---|---|---|
| AA Math Index | 91.0 | Not measured | Only Qwen3 235B A22B Instruct 2507 measured |
| AIME 2025 (Artificial Analysis run) | 91.0% | Not measured | Only Qwen3 235B A22B Instruct 2507 measured |
| Neither model has a result on the other 11 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
| Benchmark | Qwen3 235B A22B Instruct 2507 | Step 3.7 Flash | Leader |
|---|---|---|---|
| MMLU Pro (Artificial Analysis run) | 84.3% | Not measured | Only Qwen3 235B A22B Instruct 2507 measured |
| Neither model has a result on the other 17 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index06
| Benchmark | Qwen3 235B A22B Instruct 2507 | Step 3.7 Flash | Leader |
|---|---|---|---|
| Design Arena: websites | 1058 | 1196 | Step 3.7 Flash |
| Design Arena: UI components | 962 | 1177 | Step 3.7 Flash |
| Design Arena: game dev | 951 | 1158 | Step 3.7 Flash |
| Design Arena: data viz | 1070 | 1173 | Step 3.7 Flash |
| Design Arena: 3D | 994 | 1129 | Step 3.7 Flash |
| Design Arena: all code categories | 1038 | 1182 | Step 3.7 Flash |
| Design Arena: SVG | Not measured | 1084 | Only Step 3.7 Flash measured |
| Design Arena: ASCII art | Not measured | 1152 | Only Step 3.7 Flash measured |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Qwen3 235B A22B Instruct 2507 | Step 3.7 Flash | Leader |
|---|---|---|---|
| IFBench | 51.2% | 67.3% | Step 3.7 Flash |
| Long context reasoning (AA-LCR) | 72.0% | 73.7% | Step 3.7 Flash |
| DTBench | 78.4% | Not measured | Only Qwen3 235B A22B Instruct 2507 measured |
| LMCA | 23.5% | Not measured | Only Qwen3 235B A22B Instruct 2507 measured |
| Neither model has a result on the other 42 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Qwen3 235B A22B Instruct 2507 | Step 3.7 Flash | 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
Qwen3 235B A22B Instruct 2507 costs less
Repository review
50K input + 3K output tokens
Qwen3 235B A22B Instruct 2507 costs less
Agent loop
200K input + 10K output tokens
Qwen3 235B A22B Instruct 2507 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 | Qwen3 235B A22B Instruct 2507 | StepFun Step 3.7 Flash |
|---|---|---|
| Maker | Qwen | StepFun |
| Weights | Open weights | Open weights |
| Released | July 21, 2025 | May 28, 2026 |
| Context window | 262K | 262K |
| Input $/Mtok | $0.15 | $0.20 |
| Output $/Mtok | $0.75 | $1.15 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 0 | 0 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Qwen3 235B A22B Instruct 2507 better than Step 3.7 Flash for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Qwen3 235B A22B Instruct 2507 for output tokens 1.5x cheaper ($0.75 vs $1.15 per million) and work measured by SciCode or Terminal-Bench Hard (Artificial Analysis run), where it leads.
Which is cheaper, Qwen3 235B A22B Instruct 2507 or Step 3.7 Flash?
Qwen3 235B A22B Instruct 2507 costs $0.15 per million input tokens and $0.75 per million output tokens; Step 3.7 Flash costs $0.20 and $1.15. Cost per solved task is known for neither of them.
Which has the bigger context window, Qwen3 235B A22B Instruct 2507 or Step 3.7 Flash?
Neither: both take 262K tokens.
How many benchmarks is the Qwen3 235B A22B Instruct 2507 vs Step 3.7 Flash comparison based on?
14 benchmarks have a result for both models as of October 22, 2026. Qwen3 235B A22B Instruct 2507's coding index rests on 0 independent benchmarks and Step 3.7 Flash's on 0. Where only one model has a result, the page says so instead of naming a winner.