{"slug": "stepfun-releases-step-3-7-flash-a-198b-moe-vision-language-model-for-coding-and", "title": "StepFun Releases Step 3.7 Flash: A 198B MoE Vision-Language Model for Coding Agents and Search Workflows", "summary": "StepFun released Step 3.7 Flash, a 198-billion parameter Mixture-of-Experts vision-language model with native image understanding and improved tool-use reliability for agentic coding and search workflows. The model achieves 56.26% on SWE-Bench Pro and 59.55% on Terminal-Bench 2.1, with an Advisor Mode that reaches 97% of Claude Opus 4.6's coding performance at roughly one-ninth the cost. Step 3.7 Flash is available under an Apache 2.0 license with a 256k-token context window and up to 400 tokens per second throughput.", "body_md": "StepFun today released [Step 3.7 Flash](https://github.com/stepfun-ai/Step-3.7-Flash), a multimodal Mixture-of-Experts model targeting agentic use cases. It adds native vision input and improved tool-use reliability over Step 3.5 Flash.\n\n**What is Step 3.7 Flash?**\n\nStep 3.7 Flash is a **198B-parameter sparse Mixture-of-Experts (MoE) vision-language model**. It pairs a **196B-parameter language backbone** with a **1.8B-parameter vision encoder (ViT)** for native image understanding.\n\nThe model activates approximately **11B parameters per token** during inference. In MoE architectures, only a subset of “expert” sub-networks fires per forward pass — not the full network. This keeps inference compute closer to an 11B dense model while maintaining a 198B total parameter budget.\n\n**Key specs:**\n\n| Spec | Value |\n|---|---|\n| Total parameters | 198B (196B language + 1.8B ViT) |\n| Active parameters per token | ~11B |\n| Context window | 256k tokens |\n| Throughput | Up to 400 tokens/sec |\n| Reasoning levels | Low, medium, high |\n| License | Apache 2.0 |\n\n**Architecture Notes**\n\nThe vision encoder runs as a separate 1.8B ViT module. It injects image representations into the language backbone’s context. Step 3.5 Flash had no multimodal support; this is a new addition in 3.7.\n\nThree selectable reasoning depths — low, medium, and high — let developers trade latency for reasoning depth. Low is faster and cheaper; high applies more computation per response.\n\n**Agentic Coding Performance**\n\nOn **SWE-Bench Pro**, Step 3.7 Flash scores **56.26%**, up from Step 3.5 Flash’s 51.3% — a gain of roughly 5 percentage points. On **Terminal-Bench 2.1**, it scores **59.55%**, up from 53.37%.\n\nOn **SWE-MTLG** (a multi-task long-generation coding benchmark), it scores **72.42%**.\n\nCross-harness consistency on StepFun’s internal **Step-SWE-Bench**:\n\n| Scaffold | Step 3.7 Flash | Step 3.5 Flash |\n|---|---|---|\n| Hermes Agent | 67.5% | 60.0% |\n| OpenClaw | 67.0% | 47.0% |\n| KiloCode | 67.5% | 59.0% |\n| RooCode | 64.5% | 43.0% |\n| Claude Code | 71.5% | 73.0% |\n| OpenCode | 64.5% | 57.0% |\n\nStep 3.5 Flash ranged from 43% to 73% across harnesses. Step 3.7 Flash ranges from 64.5% to 71.5%. In production, coding agents often run inside heterogeneous scaffolds — each with its own prompting conventions and tool schemas. Narrower per-harness variance means more predictable behavior across different setups.\n\n**Advisor Mode**\n\nStep 3.7 Flash supports **Advisor Mode**, StepFun’s implementation of the advisor strategy described by Anthropic. The model runs the agentic loop end-to-end — calling tools, reading results, iterating — and escalates to a larger advisor model only at specific inflection points, such as planning or recovering from repeated failures. Most of the run stays at executor cost.\n\nWith Advisor Mode enabled on SWE-Bench Verified, StepFun reports Step 3.7 Flash reaches **97% of Claude Opus 4.6’s coding performance at roughly one-ninth the per-task cost** ($0.19 vs. $1.76 per task). These are StepFun’s internal figures.\n\n**Multimodal Capabilities**\n\n**Step 3.7 Flash supports two visual tool pathways:**\n\n**Visual Search Tool** — For recognition tasks where the model’s parametric knowledge is insufficient (long-tail entities, recently emerged concepts), it invokes a visual search tool to retrieve and verify. On **SimpleVQA (with Search)**, it scores **79.16%**, comparable to GPT 5.5 (79.11%) and above Kimi K2.6 (78.24%) and GLM 5V Turbo (78.20%).\n\n**Python Tool** — For fine-grained visual tasks (high-resolution images, visual probing, bounding-box analysis), it uses a code interface to crop, zoom, and draw pixels or bounding boxes. On *V* (a self-tested score with Python), it scores **95.29%**. On **HR-Bench 4K** and **HR-Bench 8K**, it scores **89.13%** and **86.34%** respectively.\n\nStepFun notes an observed behavior during testing: the model combined visual tools with non-visual tools without being explicitly trained to do so. For example, after generating frontend code, it used the GUI to render and inspect the result before iterating. StepFun describes this as emergent compositional tool use.\n\nOn **Android Daily** (long-horizon phone UI task completion), Step 3.7 Flash scores **61.87%**, ahead of Kimi K2.6 (53.36%) and GLM 5V Turbo (51.68%). Gemini 3 Flash (63.21%) leads this benchmark.\n\n**Search and Research Benchmarks**\n\nStepFun focused this model’s search design on planning, evidence filtering, and synthesis — integrating search as part of the reasoning loop rather than a separate add-on.\n\n| Benchmark | Step 3.7 Flash | Notable comparison |\n|---|---|---|\n| HLE with Tools (acc) | 47.20% | DeepSeek V4 Flash: 45.10% |\n| BrowseComp (acc) | 75.82% | Claude Opus 4.7: 79.30% |\n| DeepSearchQA (F1) | 92.82% | Kimi K2.6: 92.50% |\n| ResearchRubrics (score) | 71.68% | GPT 5.5: 61.50% |\n\nNote: The HLE with Tools score of 47.20% compares to Step 3.5 Flash’s text-only score of 35.68%. Step 3.5 Flash did not support tool-augmented evaluation on HLE.\n\n**General Agent Benchmarks**\n\n| Benchmark | Step 3.7 Flash | Description |\n|---|---|---|\n| Toolathlon | 49.51% | Multi-tool coordination |\n| ClawEval-1.1 | 67.07% | Daily autonomous task execution in realistic environments |\n| GDPval (44 occupations) | 45.8% | General professional task execution |\n| Tau2-bench Telecom | >98% | Across different reasoning difficulty tiers |\n\nOn ClawEval-1.1, Step 3.7 Flash (67.07%) leads DeepSeek V4 Flash (57.80%) and DeepSeek V4 Pro (59.80%) among the compared models.\n\n**Long-Context Performance**\n\nOn **AA-LCR** (a long-context retrieval benchmark, avg@16/acc), Step 3.7 Flash scores **63.94%**. This is comparable to DeepSeek V4 Flash (63.70%) and DeepSeek V4 Pro (66.30%).\n\n**Pricing**\n\n| Token Type | Price |\n|---|---|\n| Input (cache miss) | $0.20 / M tokens |\n| Input (cache hit) | $0.04 / M tokens |\n| Output | $1.15 / M tokens |\n\n**Marktechpost’s Visual Explainer**\n\n**Key Takeaways**\n\n- Step 3.7 Flash is a 198B sparse MoE model with 11B active params and a 256k context window.\n- Native multimodal support (images, GUIs, documents) is new — Step 3.5 Flash was text-only.\n- Advisor Mode reaches 97% of Claude Opus 4.6's SWE-Bench Verified performance at $0.19 per task vs. $1.76.\n- Cross-harness coding variance narrowed from a 43–73% range (3.5 Flash) to 64.5–71.5% (3.7 Flash).\n- Released under Apache 2.0 with BF16, FP8, NVFP4, and GGUF weights on Hugging Face.\n\nCheck out the ** Model Weights**,\n\n**and**\n\n[Repo](https://github.com/stepfun-ai/Step-3.7-Flash)**Also, feel free to follow us on**\n\n[Technical Details](https://static.stepfun.com/blog/step-3.7-flash/).**and don’t forget to join our**[Twitter](https://x.com/intent/follow?screen_name=marktechpost)\n\n**and Subscribe to**\n\n[150k+ ML SubReddit](https://www.reddit.com/r/machinelearningnews/)**. Wait! are you on telegram?**\n\n[our Newsletter](https://www.aidevsignals.com/)\n\n[now you can join us on telegram as well.](https://t.me/machinelearningresearchnews)Need to partner with us for promoting your GitHub Repo OR Hugging Face Page OR Product Release OR Webinar etc.? [Connect with us](https://forms.gle/wbash1wF6efRj8G58)", "url": "https://wpnews.pro/news/stepfun-releases-step-3-7-flash-a-198b-moe-vision-language-model-for-coding-and", "canonical_source": "https://www.marktechpost.com/2026/05/29/stepfun-releases-step-3-7-flash-a-198b-moe-vision-language-model-for-coding-agents-and-search-workflows/", "published_at": "2026-05-29 21:25:05+00:00", "updated_at": "2026-05-29 21:31:03.372443+00:00", "lang": "en", "topics": ["large-language-models", "computer-vision", "ai-agents", "ai-research", "ai-products"], "entities": ["StepFun", "Step 3.7 Flash", "Step 3.5 Flash", "Mixture-of-Experts", "ViT", "Apache 2.0"], "alternates": {"html": "https://wpnews.pro/news/stepfun-releases-step-3-7-flash-a-198b-moe-vision-language-model-for-coding-and", "markdown": "https://wpnews.pro/news/stepfun-releases-step-3-7-flash-a-198b-moe-vision-language-model-for-coding-and.md", "text": "https://wpnews.pro/news/stepfun-releases-step-3-7-flash-a-198b-moe-vision-language-model-for-coding-and.txt", "jsonld": "https://wpnews.pro/news/stepfun-releases-step-3-7-flash-a-198b-moe-vision-language-model-for-coding-and.jsonld"}}