Capability
51.1/100
field median 58.8
#126 of 228 ranked models
Step 3.7 Flash
Released May 29, 2026 — see all recent releases
Decision reading
Step 3.7 Flash scores 51.1 out of 100 and ranks #126 of 228. This profile shows 11 source-displayable benchmark rows; its strongest eligible category is Coding at #75. API pricing is $0.2 input and $1.15 output per million tokens.
Data as of August 30, 2026 · How the score is built
Strongest published evidence
Coding ranks #75. Particularly well-suited for software development and code generation tasks.
Validate before choosing
11 published rows leave some tracked benchmark slots empty. Agentic is its lowest eligible category at #84.
Decision snapshot
Each value carries a field reference instead of floating alone. Markers compare this model with the current ranked and priced catalog; they are not absolute quality thresholds.
Price
$0.20input / $1.15 output
input median $1
blended $0.67
Speed
87tok/s
field median 86 tok/s
First token 25.81 s
Context
256Ktokens
field median 256,000
Reported for this model; direct source link not stored
Capability shape
Each axis shows percentile within that category’s eligible cohort. The comparison outline is the median of the six nearest public-score peers; a collapsed vertex means the category is not rank-eligible.
Step 3.7 Flash category percentile values
- Agentic40th percentile
- Coding49th percentile
- ReasoningNot eligible
- KnowledgeNot eligible
- MathNot eligible
- MultilingualNot eligible
- MultimodalNot eligible
- Instruction followingNot eligible
The dashed outline is median of 6 nearest peers.
Eligible category ranks
- Agentic#84/140
- Coding#75/146
- ReasoningNot ranked
- KnowledgeNot ranked
- MathNot ranked
- MultilingualNot ranked
- MultimodalNot ranked
- Inst. FollowingNot ranked
Top decileTop quartileMid-fieldNot eligible
What it costs to get this score
Published API price against the public score. The x-axis uses a log scale; the dashed path marks models that are not beaten by a cheaper, higher-scoring option. Price uses average of published input and output rates.
The chart opens on the current model. Scroll horizontally to inspect the full price axis.
Current modelStep 3.7 Flash · 51.0 score · $0.67 blended per million tokens
Horizontal: blended price per million tokens, log scale · Vertical: public score
How much of this is verified
Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.
- Agentic7/7 verified
- Coding2/2 verified
- ReasoningNot measured
- KnowledgeNot measured
- MathNot measured
- MultilingualNot measured
- Multimodal2/2 verified
- Inst. FollowingNot measured
Verified sourceProvisionalNot measured
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- Not published
- Context window
- 256K
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- Not sourced yet
- Output modalities
- Not sourced yet
- Parameters
- Not sourced yet
- Availability
- Not sourced yet
- Cloud regions
- Not tracked yet
- Lifecycle
- Current
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Not documented in the pricing record
- Self-host
- Open weights available; hardware estimate not sourced
- Rate limits
- Not tracked yet
Deployment options
Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.
Published weights are available, but BenchLM does not yet have a sourced parameter and VRAM profile for this exact model. Hardware cost estimates stay unavailable until that sizing record is complete.
Estimate VRAM from known parametersCategory score record
Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticRank #84 of 140Percentile 40thWeight 22%7 benchmarksVerified | 45.6 | #84 of 140 | 40th | 22% | 7 benchmarks | Verified |
| CodingRank #75 of 146Percentile 49thWeight 20%2 benchmarksVerified | 50.8 | #75 of 146 | 49th | 20% | 2 benchmarks | Verified |
| ReasoningWeight 17%0 benchmarksNot measured | Not measured | Not ranked | Not available | 17% | 0 benchmarks | Not measured |
| KnowledgeWeight 12%0 benchmarksNot measured | Not measured | Not ranked | Not available | 12% | 0 benchmarks | Not measured |
| MathWeight 5%0 benchmarksNot measured | Not measured | Not ranked | Not available | 5% | 0 benchmarks | Not measured |
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | Not ranked | Not available | 7% | 0 benchmarks | Not measured |
| MultimodalWeight 12%2 benchmarksVerified | Score pending | Not ranked | Not available | 12% | 2 benchmarks | Verified |
| Inst. FollowingWeight 5%0 benchmarksNot measured | Not measured | Not ranked | Not available | 5% | 0 benchmarks | Not measured |
Benchmark ledger
Coding opens by default. The marker compares each value with the best source-verified result in the catalog; provisional leaders do not set the reference. Expand the remaining categories for every published row.
Coding2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| SWE-bench Pro | Score56.3% | Versus best verified row Best verified: Claude Mythos 5 · 80.3% | Gap24 behind | WeightWeighted 10% | Provider exact |
| Terminal-Bench 2.0 | Score59.5% | Versus best verified row Best verified: GPT-5.6 Sol · 91.9% | Gap32.4 behind | WeightDisplay only | Provider exact |
Agentic7 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| Terminal-Bench 2.0 | Score59.5% | Versus best verified row Best verified: GPT-5.6 Sol · 91.9% | Gap32.4 behind | WeightWeighted 38% | Provider exact |
| BrowseComp | Score75.8% | Versus best verified row Best verified: GPT-5.6 Sol · 92.2% | Gap16.4 behind | WeightWeighted 28% | Provider exact |
| DeepSearchQA | Score92.8% | Versus best verified row Best verified: Claude Opus 5 · 95.0% | Gap2.2 behind | WeightDisplay only | Provider exact |
| Toolathlon | Score49.5% | Versus best verified row Best verified: Muse Spark 1.1 · 75.6% | Gap26.1 behind | WeightDisplay only | Provider exact |
| Claw-Eval | Score67.1% | Versus best verified row Best verified: Ornith-1.5-397B · 81.4% | Gap14.3 behind | WeightDisplay only | Provider exact |
| HLE w/ toolsHumanity's Last Exam with tools | Score47.2% | Versus best verified row Best verified: Claude Opus 5 · 64.7% | Gap17.5 behind | WeightDisplay only | Provider exact |
| Gert LabsGert Labs Composite Game Benchmark | Score51.57% | Versus best verified row Best verified: Claude Opus 4.8 · 72.97% | Gap21.4 behind | WeightDisplay only | Benchmark exact |
Multimodal2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| SimpleVQA | Score79.2% | Versus best verified row Best verified: Qwen3.7 Plus · 81.7% | Gap2.5 behind | WeightDisplay only | Provider exact |
| V* | Score95.3% | Versus best verified row Best verified: Kimi K2.6 · 96.9% | Gap1.6 behind | WeightDisplay only | Provider exact |
Lineage
The sequence follows explicit supersedes links. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Jan 20, 2026
Step 3.5 FlashNot publicly ranked · $0.1 / $0.3
May 29, 2026 · you are here
Step 3.7 FlashScore 51.0 · $0.2 / $1.15
Base entry
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
Step 3.7 Flash ranks #126 of 228 on the public leaderboard with a score of 51.05/100. It does not yet have enough sourced coverage for a verified position.
Step 3.7 Flash is a open weight model with a 256K context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
Official exact-value snapshot from StepFun's May 29, 2026 Step 3.7 Flash launch page and model card. BenchLM maps only the directly comparable public values; Step-SWE-Bench, ResearchRubrics, HR-Bench, VisualProbe, and the separate Hugging Face HLE 48.1 row remain out of schema or setup-ambiguous for now.
Its explicit predecessor is Step 3.5 Flash. 11 of 408 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Its strongest eligible category is Coding at #75, while its lowest eligible position is Agentic at #84. particularly well-suited for software development and code generation tasks.
Radar
Full release historyStep 3.7 Flash release history
Step 3.7 Flash released
StepFun · Model release
Source confirmed
Radar confirmed these at the source. Start the free Radar Brief
Frequently asked questions
How does Step 3.7 Flash perform overall in AI benchmarks?
Step 3.7 Flash ranks #126 out of 228 models on the public BenchAlign leaderboard, with a score of 51.05/100. Its evidence status is Estimated, and this profile shows 11 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.
Is Step 3.7 Flash good for coding and programming?
Step 3.7 Flash ranks #75 out of 146 eligible models for coding and programming, with a public category score of 50.8/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Step 3.7 Flash good for agentic tool use and computer tasks?
Step 3.7 Flash ranks #84 out of 140 eligible models for agentic tool use and computer tasks, with a public category score of 45.6/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Step 3.7 Flash good for multimodal and grounded tasks?
Step 3.7 Flash has source-displayable benchmark coverage for multimodal and grounded tasks, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Step 3.7 Flash open source?
Step 3.7 Flash is an open-weight model from StepFun. Its weights can be downloaded for local or hosted deployment, subject to the published license. Open weight does not automatically mean open source: training data and training code may remain private, and commercial restrictions can still apply.
Does Step 3.7 Flash have full benchmark coverage on BenchLM?
No. Step 3.7 Flash currently has 29 source-displayable rows across 408 tracked benchmark slots. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of Step 3.7 Flash?
Step 3.7 Flash has a reported context window of 256K in the exact-model catalog record. The value stays visible, but the profile marks its source link as unavailable instead of presenting it as directly documented. Maximum output length remains separate because providers often publish a different limit.
Compare Step 3.7 Flash with every tracked model399 comparisons
Last updated August 30, 2026. Runtime fields remain blank until a sourced snapshot exists.
Watch Step 3.7 Flash in the weekly brief
Get one weekly email when material rank, price, availability, or benchmark evidence changes are worth revisiting.
Read a sample issueJoin 2,000+ readers.