BazaarLinkBazaarLink
Sign in
S

Step 3.5 Flash — Currently unavailable

Currently unavailableBazaarLink no longer offers this model. This page remains available as a historical reference.

Step 3.5 Flash is StepFun's most capable open-source foundation model. Built on a sparse Mixture of Experts (MoE) architecture, it selectively activates only 11B of its 196B parameters per token....

ProviderSstepfun
ReleasedJanuary 2026
Model IDstepfun/step-3.5-flash

Technical specifications

Context window
262K tokens
Reasoning
Input
Text
Output
Text

Independent benchmarks

Scores below come from Artificial Analysis, an independent third party. BazaarLink does not participate in the testing.

Intelligence 17
Coding
Math
MMLU
GPQA 83
This modelCategory leader
Overall intelligence
#217out of 638 models
Aggregated across academic benchmarks
Time to first token
#215out of 329 models
AA 全球 p50:3.33 秒
Intelligence
17
Mid-tier
This model
17
Category leader
53
Median
12
GPQA
83%
This model
83%
Category leader
94%
Median
69%
HLE
23%
This model
23%
Category leader
53%
Median
7%
SciCode
39%
This model
39%
Category leader
60%
Median
34%
IFBench
67%
This model
67%
Category leader
83%
Median
44%
τ²-Bench Telecom
87%
This model
87%
Category leader
99%
Median
46%
AA-LCR
54%
This model
54%
Category leader
76%
Median
40%
Terminal-Bench Hard
33%
This model
33%
Category leader
66%
Median
14%

Top 5 — CodingCoding Index leaderboard

2Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)?/claude-fable-5-1-xhigh81$10 / $50
3Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)?/claude-fable-5-1-high79$10 / $50
4GPT-5.6 Sol (xhigh)openai/gpt-5-6-sol-xhigh78$4 / $20
AASource: Artificial Analysis · Independent third-party evaluation; not influenced by BazaarLinkView full evaluation on AA →
← All Models

Frequently Asked Questions

Can I still use Step 3.5 Flash through BazaarLink?

No. Step 3.5 Flash is currently unavailable through BazaarLink. See the active alternatives linked on this page.

Relay integrity checks

Tested endpoints
1 runs
Independent hosts
1
Model-family fingerprint match
0%
Anomalies detected
1 runs
Data range
past 90 days

BazaarLink Probe verifies endpoints that claim to serve this model and flags behavior that differs from the reference baseline.

View the full integrity report →

Related models and links

Related models and links

Qwen3.8 MaxDeepseek V3.2Glm 5.1Deepseek V4 FlashGlm 4.7Minimax M2.7
ModelsAPI documentationAPI latency probe
Support
Support
Hi! How can we help you?
Send a message and we'll get back to you soon.