StepFun
·Released on May 28, 2026Step 3.7 Flash API Benchmarks, Pricing & Provider Data
Compare Step 3.7 Flash with another model
Choose a model to open its comparison page.
LLM
Step 3.7 Flash benchmark, API pricing, and provider data cover undefined API provider} other undefined API providers}}, with prices starting at $0.00005/request. Step 3.7 Flash free API options are available from undefined provider} other undefined providers}}. The page also shows measured API speed and first-token latency.
Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model. It pairs a 196B-parameter language backbone with a vision encoder for native image and video understanding, acti...
Quality
#75of 112
53.0
LMSpeed score
Speed
#31of 80
148char/s
22.50 s
Cost
#35of 186
$0.00005/ 1M · 8:1 in:out
$0.000008 in · $0.000042 out
Category Performance
Observed-capability estimates with uncertainty reported separately from independent benchmark families.
- Coverage
- 6 / 8
- Methodology
- V3.0
#1Multimodal59Provisional1/4 Measured dimensions
80% interval: 44.5–73.5
perception ocr59
simplevqa · simple_vqa / v_star · v_star
document spatialPrior only
visual reasoningPrior only
video actionPrior only
#2Agents53.3RatedGlobal rank #284/4 Measured dimensions
80% interval: 47.6–59.0
planning51.3
deep_planning · gert_labs
tool use58.6
tau · tau2_bench / toolathlon · toolathlon
environment execution51.2
browsecomp · browse_comp / deep_search_qa · deep_search_qa / gdpval_aa · benchlm_agentic_gdpval_aa / terminalbench · benchlm_agentic_terminal_bench2
recovery reliability52.1
apex_agents · apex_agents_aa / claw_eval · claw_eval
#3Reasoning51.8Estimated2/4 Measured dimensions
80% interval: 41.0–62.6
abstract logicPrior only
scientific causal52.7
critpt · critpt / gpqa · gpqa / hle · hle
multistep constraintsPrior only
evidence verification50.9
lcr · lcr
#4Instruction following51.7Provisional1/4 Measured dimensions
80% interval: 35.7–67.7
constraint followingPrior only
structured outputPrior only
novel instruction generalization51.7
ifbench · aa_if_bench
long multiturn instructionPrior only
#5Coding48.7Estimated3/4 Measured dimensions
80% interval: 39.4–58.0
code generation51.7
scicode · scicode
repository engineering48.2
swe_pro · swe_pro
debugging testingPrior only
tooling quality46.2
terminalbench · benchlm_coding_terminal_bench2
#6Knowledge42.1Provisional1/4 Measured dimensions
80% interval: 26.1–58.1
broad knowledge42.1
aa_omniscience · aa_omniscience_index
professional knowledgePrior only
factualityPrior only
retrieval open bookPrior only
No data:MathMultilingual
Specifications
Input and output token limits for this model, plus how it ranks on long-context understanding.
INPUT
262.1Ktokens
≈ 314.6 pages of text
OUTPUT
230.4Ktokens
8K128K1M4M
262.1K
Features
Technical Details
- Input
- Output
- Released
- May 2026
- Tokenizer
- Other
- Architecture
- text+image+video->text
- Moderated
- No
- Supported parameters
- frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
Rankings
Excels at
It's decent at
Falls behind in
Detailed scores
Updated: Sep 13, 2026Overall
undefined metric} other undefined metrics}}
Overall score53.0#75 / 112
Overall
undefined metric} other undefined metrics}}
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Speed & latency
undefined metric} other undefined metrics}}
Output speed133.3 tok/s#31 / 80Time to first token1.58 s#44 / 80
Speed & latency
undefined metric} other undefined metrics}}
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Output speed
LMSpeed rank#31
Score133.3 tok/s
Sourceartificialanalysis.ai
UpdatedSep 13, 2026
Confidence4
Time to first token
LMSpeed rank#44
Score1.58 s
Sourceartificialanalysis.ai
UpdatedSep 13, 2026
Confidence4
Pricing
undefined metric} other undefined metrics}}
Input price$0.200/M#35 / 186Output price$1.15/M#49 / 186
Pricing
undefined metric} other undefined metrics}}
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Agents
V3.0undefined metric} other undefined metrics}} · Rated
Score53.3#2880% interval47.6–59.04/4 Measured dimensionsAgentic score52.3#51 / 77Terminal-Bench 2.059.5#32 / 53
Agents
V3.0undefined metric} other undefined metrics}} · Rated
Dimensions and evidence
Planning & decomposition51.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
deep_planning · gert_labs · z 0.18 · q 1.00
Tool use58.6
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
tau · tau2_bench · z 1.32 · q 1.00
toolathlon · toolathlon · z 0.15 · q 1.00
Environment & long-horizon execution51.2
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
browsecomp · browse_comp · z -0.48 · q 1.00
deep_search_qa · deep_search_qa · z 0.85 · q 1.00
gdpval_aa · benchlm_agentic_gdpval_aa · z -0.37 · q 1.00
terminalbench · benchlm_agentic_terminal_bench2 · z -0.27 · q 1.00
Recovery & completion reliability52.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
apex_agents · apex_agents_aa · z -0.81 · q 1.00
claw_eval · claw_eval · z 1.27 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Coding
V3.0undefined metric} other undefined metrics}} · Estimated
Score48.780% interval39.4–58.03/4 Measured dimensionsSciCode43.9%#50 / 89Coding score41.2#74 / 87
Coding
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Code generation51.7
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
scicode · scicode · z 0.06 · q 1.00
Repository engineering48.2
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
swe_pro · swe_pro · z -0.14 · q 1.00
Debugging & testingPrior only
Tool-assisted development & quality46.2
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
terminalbench · benchlm_coding_terminal_bench2 · z -0.49 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
SciCodeV3 evidence
LMSpeed rank#50
Score43.9%
Sourceartificialanalysis.ai
UpdatedSep 13, 2026
Confidence4
Reasoning
V3.0undefined metric} other undefined metrics}} · Estimated
Score51.880% interval41.0–62.62/4 Measured dimensionsGPQA80.9%#97 / 218HLE21.4%#83 / 216
Reasoning
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Abstract logicPrior only
Scientific & causal reasoning52.7
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
critpt · critpt · z -0.12 · q 1.00
gpqa · gpqa · z 0.17 · q 1.00
hle · hle · z 0.29 · q 1.00
Multi-step constraintsPrior only
Evidence integration & verification50.9
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
lcr · lcr · z -0.23 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Knowledge
V3.0undefined metric} other undefined metrics}} · Provisional
Score42.180% interval26.1–58.11/4 Measured dimensionsArtificial Analysis Intelligence Index30.9#46 / 117AA-GPQA Diamond80.9#77 / 113
Knowledge
V3.0undefined metric} other undefined metrics}} · Provisional
Dimensions and evidence
Broad knowledge42.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
aa_omniscience · aa_omniscience_index · z -1.14 · q 1.00
Professional knowledgePrior only
FactualityPrior only
Retrieval & open-book usePrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Artificial Analysis Intelligence IndexV3 evidence
LMSpeed rank#46
Score30.9
Sourcebenchlm.ai
UpdatedAug 20, 2026
Confidence1
AA-Omniscience Hallucination Rate
LMSpeed rank#30
Score85.0
Sourcebenchlm.ai
UpdatedAug 20, 2026
Confidence1
Math
V3.0undefined metric} other undefined metrics}} · No data
No data80% interval30.8–69.20/4 Measured dimensions
Math
V3.0undefined metric} other undefined metrics}} · No data
Dimensions and evidence
Foundational mathPrior only
Competition mathPrior only
Advanced proofsPrior only
Applied & tool-assisted mathPrior only
Multilingual
V3.0undefined metric} other undefined metrics}} · No data
No data80% interval30.8–69.20/4 Measured dimensions
Multilingual
V3.0undefined metric} other undefined metrics}} · No data
Dimensions and evidence
Cross-language understandingPrior only
Multilingual generationPrior only
Reasoning transferPrior only
Low-resource robustnessPrior only
Multimodal
V3.0undefined metric} other undefined metrics}} · Provisional
Score5980% interval44.5–73.51/4 Measured dimensionsSimpleVQA79.2#2 / 8V*95.3#4 / 11
Multimodal
V3.0undefined metric} other undefined metrics}} · Provisional
Dimensions and evidence
Perception & OCR59
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
simplevqa · simple_vqa · z 1.04 · q 1.00
v_star · v_star · z 0.52 · q 1.00
Document & spatial understandingPrior only
Visual reasoningPrior only
Video & grounded actionPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Instruction following
V3.0undefined metric} other undefined metrics}} · Provisional
Score51.780% interval35.7–67.71/4 Measured dimensionsAA-IFBench67.3#43 / 84
Instruction following
V3.0undefined metric} other undefined metrics}} · Provisional
Dimensions and evidence
Constraint followingPrior only
Structured outputPrior only
Novel-instruction generalization51.7
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
ifbench · aa_if_bench · z -0.07 · q 1.00
Multi-turn & long instructionsPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
OpenRouter endpoints
3 endpointsThird-party OpenRouter endpoint data, shown separately from LMSpeed measurements. Some 30-minute live performance fields only appear after syncing with an OpenRouter API key.
| Provider endpoint | Input | Output | 1d uptime | 30m latency | 30m throughput | Context / output |
|---|---|---|---|---|---|---|
DeepInfra deepinfra | $0.160/M | $0.920/M | 99.9% | — | — | undefined tokens / undefined tokens |
StepFun stepfun/fp8 | $0.200/M | $1.15/M | 99.4% | — | — | undefined tokens / undefined tokens |
Novita novita/fp8 | $0.200/M | $1.15/M | 99.4% | — | — | undefined tokens / undefined tokens |
Pricing Comparison
Compare Step 3.7 Flash API pricing across 48 providers. Prices range from $0.00005/request to $75.00/M. CM-API 公益站 offers the lowest rate at $0.00005/request. 9 providers offer free API credits or a free tier.
| Provider | Health | Model Variant | Group | Input ($/M) | Output ($/M) | Speed (t/s) | First token | Audit |
|---|---|---|---|---|---|---|---|---|
L1 0% | step-3.7-flash | nvidia | -100%$0.0002/M | -100%$0.0011/M | 71.7 t/s | 17.60 s | — | |
L1 100% L2 100% | step-3.7-flash | Archived | -8%$0.185/M Cache read$0.037/M | -3%$1.11/M | 36.5 t/s | 2.76 s | — | |
L1 100% | step-3.7-flash | default | -26%$0.148/M | -23%$0.887/M | — | — | — | |
L1 100% | step-3.7-flash | default | -93%$0.014/M | -93%$0.079/M | — | — | — | |
L1 100% | step-3.7-flash | default | $7.50/M | $7.50/M | — | — | — | |
L1 100% | stepfun-ai/Step-3.7-Flash | default | $7.50/M | $7.50/M | — | — | — | |
L1 100% L2 0% | step-3.7-flash | nvidia | $75.00/M | $75.00/M | — | — | — | |
L1 100% | Step-3.7-Flash | free | $0.00005/request | - | — | — | — | |
L1 100% | stepfun/step-3.7-flash:free | free | $0.00005/request | - | — | — | — | |
L1 100% | daipai/step-3.7-flash | 按次福利模型 | $0.0020/request | - | — | — | — | |
兔子API Free | L1 99% | step-3.7-flash-free | default | Free | Free | — | — | — |
L1 100% | stepfun-ai/Step-3.7-Flash | default | $0.479/request | - | — | — | — | |
L1 100% | step-3.7-flash | 免费 | -97%$0.0067/M Cache read$0.0014/M | -96%$0.041/M | — | — | — | |
L1 100% | step-3.7-flash | default | $0.675/M Cache read$0.135/M | $4.05/M | — | — | — | |
Moyanjdc API Free | L1 99% | step-3.7-flash | default | Free | Free | — | — | — |
L1 100% | stepfun-ai/step-3.7-flash | default | $75.00/M | $75.00/M | — | — | — | |
L1 99% | stepfun-ai/step-3.7-flash | 0倍倍率分组 | -93%$0.014/M | -96%$0.041/M | — | — | — | |
L1 0% | stepfun-ai/step-3.7-flash | 2api | $3.00/M | $3.00/M | — | — | — | |
L1 61% | stepfun-ai/step-3.7-flash | default | $75.00/M | $75.00/M | — | — | — | |
L1 100% | step-3.7-flash | nvidia专区 | -79%$0.041/M Cache read$0.0082/M | -79%$0.236/M | — | — | — |
Showing 20 model IDs of 31.
Alternatives & Similar Models
GLM-5.1
glm-5-1
Zhipu GLM-5.1 is a next-generation GLM model aimed at frontier reasoning, coding, and bilingual agent applications.
62 shared providers
DeepSeek V4 Flash
deepseek-v4-flash
DeepSeek V4 Flash is a fast, cost-efficient language model in the DeepSeek V4 family, optimized for low-latency chat, coding assistance, and high-throughput API workloads while retaining strong reasoning quality.
61 shared providers
MiniMax M2.7
minimax-m2-7
MiniMax M2.7 is a high-tier M2-series model tuned for complex reasoning, long-context dialogue, and production-grade API workloads.
61 shared providers
DeepSeek V4 Pro
deepseek-v4-pro
DeepSeek V4 Pro is the professional-tier DeepSeek V4 model, targeting frontier reasoning, coding, and agent workflows with maximum capability.
60 shared providers
MiniMax M3
minimax-m3
MiniMax M3 is MiniMax next-generation large language model, designed for advanced reasoning, long-context understanding, and high-quality multilingual dialogue.
60 shared providers
Kimi K2.5
kimi-k2-5
Moonshot Kimi K2.5 is an open-weight multimodal agent model with native vision and text input, strong coding performance, and a 256K context window.
59 shared providers
Frequently Asked Questions
- What benchmark data does Step 3.7 Flash include?
- LMSpeed shows Step 3.7 Flash benchmark context, API price, output speed, first-token latency, and provider data across 57 providers when those signals are available.
- What is the Step 3.7 Flash API price?
- Step 3.7 Flash has pricing from undefined provider} other undefined providers}}, ranging from $0.00005/request to $75.00/M. CM-API 公益站 has the lowest listed price.
- What does the Step 3.7 Flash API pricing table include?
- The Step 3.7 Flash API pricing table compares 57 providers by input price, output price, free tier status, speed, first-token latency, and recent health data when available.
- Which provider has the cheapest Step 3.7 Flash API pricing?
- CM-API 公益站 currently has the lowest listed Step 3.7 Flash price at $0.00005/request across undefined provider} other undefined providers}}.
- Can I compare Step 3.7 Flash API price and speed together?
- Yes. LMSpeed shows Step 3.7 Flash API price, output speed, first-token latency, and provider health on the same page so you can compare cost and performance together.
- Is Step 3.7 Flash API free?
- Yes, Step 3.7 Flash free API options are available through 9 providerundefined other undefined} on LMSpeed, including Zero API, 初叶🍂Furry API, 兔子API, WSocket AI, WSocket AI. These providers offer free API credits or a free tier with no per-token charges.
- Where can I get Step 3.7 Flash free API access?
- LMSpeed currently lists 9 free API providerundefined other undefined} for Step 3.7 Flash: Zero API, 初叶🍂Furry API, 兔子API, WSocket AI, WSocket AI. Check each provider row before using it because free tier limits can change.