Step 3.7 Flash vs Pareto 26.10 Preview
Step 3.7 Flash costs 2.8x less per output token than Pareto 26.10 Preview ($1.15 vs $3.20 per million). Updated November 14, 2026; 1 benchmarks have a result for both.
Step 3.7 Flash or Pareto 26.10 Preview: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick Step 3.7 Flash for output tokens 2.8x cheaper ($1.15 vs $3.20 per million) and work measured by GameBench 2, where it leads.
- Shared results
- 1
- Step 3.7 Flash only
- 19
- Pareto 26.10 Preview only
- 0
- Like-for-like categories
- 1 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
Web design
Design Arena: websites
No call
Only Step 3.7 Flash has a Design Arena: websites result, 1186.
Directional only
Long documents
Listed context window
Pareto 26.10 Preview
Pareto 26.10 Preview takes 1.05M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Step 3.7 Flash
Step 3.7 Flash $0.0008, Pareto 26.10 Preview $0.0024 per turn.
List prices
At a glance
- Coding index
- Step 3.7 Flash not benchmarked, Pareto 26.10 Preview not benchmarked.
- Widest category
- Coding, 58.9 points apart on average over 1 shared benchmark.
- Widest single gap
- GameBench 2: Step 3.7 Flash 29.4 against Pareto 26.10 Preview 88.3, 58.9 points apart on a 0 to 100 scale.
- Output price
- Step 3.7 Flash $1.15 against Pareto 26.10 Preview $3.20 per million output tokens, 2.8x apart.
- Context window
- Step 3.7 Flash takes 262K tokens, Pareto 26.10 Preview 1.05M.
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 | Step 3.7 Flash | Bars | Pareto 26.10 Preview | Ties |
|---|---|---|---|---|
| Coding | 0 | 1 | ||
| All shared boards | 0 | 1 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index01
| Benchmark | Step 3.7 Flash | Pareto 26.10 Preview | Leader |
|---|---|---|---|
| AA Coding Index | 39.6 | Not measured | Only Step 3.7 Flash measured |
| SciCode | 43.9% | Not measured | Only Step 3.7 Flash measured |
| Terminal-Bench Hard (Artificial Analysis run) | 35.6% | Not measured | Only Step 3.7 Flash measured |
| ALE-Bench | 694.1 | Not measured | Only Step 3.7 Flash measured |
| GameBench 2 | 29.4 | 88.3 | Pareto 26.10 Preview |
| Neither model has a result on the other 47 boards here | |||
IntelligenceNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Step 3.7 Flash | Pareto 26.10 Preview | Leader |
|---|---|---|---|
| Neither model has a result on 2 boards here | |||
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
| Benchmark | Step 3.7 Flash | Pareto 26.10 Preview | Leader |
|---|---|---|---|
| τ²-Bench Telecom | 98.5% | Not measured | Only Step 3.7 Flash measured |
| Neither model has a result on the other 43 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | Step 3.7 Flash | Pareto 26.10 Preview | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 80.9% | Not measured | Only Step 3.7 Flash measured |
| Humanity's Last Exam (Artificial Analysis run) | 21.4% | Not measured | Only Step 3.7 Flash measured |
| Neither model has a result on the other 24 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
| Benchmark | Step 3.7 Flash | Pareto 26.10 Preview | Leader |
|---|---|---|---|
| MathArena | 19.0% | Not measured | Only Step 3.7 Flash measured |
| Neither model has a result on the other 13 boards here | |||
KnowledgeNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Step 3.7 Flash | Pareto 26.10 Preview | Leader |
|---|---|---|---|
| Neither model has a result on 21 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
| Benchmark | Step 3.7 Flash | Pareto 26.10 Preview | Leader |
|---|---|---|---|
| Design Arena: websites | 1186 | Not measured | Only Step 3.7 Flash measured |
| Design Arena: UI components | 1167 | Not measured | Only Step 3.7 Flash measured |
| Design Arena: game dev | 1142 | Not measured | Only Step 3.7 Flash measured |
| Design Arena: data viz | 1171 | Not measured | Only Step 3.7 Flash measured |
| Design Arena: 3D | 1108 | Not measured | Only Step 3.7 Flash measured |
| Design Arena: all code categories | 1169 | Not measured | Only Step 3.7 Flash measured |
| Design Arena: SVG | 1074 | Not measured | Only Step 3.7 Flash measured |
| Design Arena: ASCII art | 1142 | Not measured | Only Step 3.7 Flash measured |
| Neither model has a result on the other 1 board here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index3 boards, none shared
| Benchmark | Step 3.7 Flash | Pareto 26.10 Preview | Leader |
|---|---|---|---|
| IFBench | 67.3% | Not measured | Only Step 3.7 Flash measured |
| Long context reasoning (AA-LCR) | 73.7% | Not measured | Only Step 3.7 Flash measured |
| GDPval-AA | 864 | Not measured | Only Step 3.7 Flash measured |
| Neither model has a result on the other 46 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Step 3.7 Flash | Pareto 26.10 Preview | 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
Step 3.7 Flash costs less
Repository review
50K input + 3K output tokens
Step 3.7 Flash costs less
Agent loop
200K input + 10K output tokens
Step 3.7 Flash 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 | StepFun Step 3.7 Flash | Pareto 26.10 Preview |
|---|---|---|
| Maker | StepFun | Unbiased |
| Weights | Open weights | No public weights listed |
| Released | May 28, 2026 | October 1, 2026 |
| Context window | 262K | 1.05M |
| Input $/Mtok | $0.20 | $0.80 |
| Output $/Mtok | $1.15 | $3.20 |
| 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 Step 3.7 Flash better than Pareto 26.10 Preview for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Step 3.7 Flash for output tokens 2.8x cheaper ($1.15 vs $3.20 per million) and work measured by GameBench 2, where it leads.
Which is cheaper, Step 3.7 Flash or Pareto 26.10 Preview?
Step 3.7 Flash costs $0.20 per million input tokens and $1.15 per million output tokens; Pareto 26.10 Preview costs $0.80 and $3.20. Cost per solved task is known for neither of them.
Which has the bigger context window, Step 3.7 Flash or Pareto 26.10 Preview?
Pareto 26.10 Preview, with 1.05M tokens against 262K.
How many benchmarks is the Step 3.7 Flash vs Pareto 26.10 Preview comparison based on?
1 benchmarks have a result for both models as of November 14, 2026. Step 3.7 Flash's coding index rests on 0 independent benchmarks and Pareto 26.10 Preview's on 0. Where only one model has a result, the page says so instead of naming a winner.