Head to head · Open-weight models you run yourself · October 2026 research run
Gemma 4 12B IT vs Step 3.7 Flash
Gemma 4 12B IT scores 72.3 (BB) on agent readiness against Step 3.7 Flash's 59.8 (C), and leads in 3 of 4 scored categories. Both do open-weight models you run yourself.
Best open-weight models you download and run yourself for AI agents · All 6359 open weights comparisons
Which one, for what
Best for Agents on a 12 GB or larger GPU or Mac that want a permissively licensed tool-calling model with images, audio and long context
Ahead on
- Reach, 95 against 70
- Risk, 70 against 45
- Documentation and upkeep, 85 against 39
Also in its favour
- Agent-ready, a grade of BB or better
Watch for
No safety evaluation figures published
Best for Teams with a GPU server or a provider account that want an open, long-context agent model with vision
No category where it leads by five points or more, and no fact that sets it apart.
Watch for
About 131 GB to run at 4 bits with images, over a 128 GB machine's share
Why the edge
Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.
Documentation and upkeep Gemma 4 12B IT +46
Gemma 4 12B IT 85
Model card 20, all seven items are there. Training data 10, a general description by type with a January 2025 cut-off and the filtering, but no sources, mix or sizes. Last change 20, Google merged a fix for 210 glitch tokens in tokenizer.json on 30 October 2026, two days before 1 November. Responsiveness 20, Google replied to 18 of 34 questions and reports, the last on 2026-07-31, re-counted on 1 November. Version policy 10, earlier Gemma generations keep their own repositories and gemma.ai has a dated releases page. File health 5, the files agree with the card but a report about them waits without a reply.
Step 3.7 Flash 39
Model card 14, Intended use, the architecture, the parameters, the context and the language (in the metadata) are on the card. Training data 0, Nothing on the card about the training data. Last change 10, the last commit, on 3 June 2026, fixed the modelling code, 160 days before this check and within 180. Responsiveness 0, none of the 11 most recent discussions that aren't pull requests (29 May to 24 June 2026) has a reply from a stepfun-ai member, counted on 10 November by author flags and dates with a script, text not read. Version policy 5, Step-3.5-Flash stays in its own repositories and 3.7 came out in new ones, with no dated note on the older cards that we read. File health 10, Config, template and card agree on the architecture and the context, and June commits fixed the modelling code. 14 + 0 + 10 + 0 + 5 + 10 = 39.
Reach Gemma 4 12B IT +25
Gemma 4 12B IT 95
Memory 30, Google's own 4-bit QAT GGUF with its projector is 7.15 GB, which with a fifth on top is 8.6 GB and fits in 16 GB or less. Runtimes 25, released support in Transformers, vLLM, SGLang, llama.cpp, Ollama and LM Studio, six of six, all for the separate gemma4_unified architecture. Builds 15, Google publishes a Q4_0 GGUF and a w4a16 build, read as two kinds as for the 31B. Terms 15, plain Apache 2.0 with nothing added. Download 10, no account or gate. Hosted 0, the Gemini API serves only the two largest sizes and no inference provider lists this one.
Step 3.7 Flash 70
Memory 5, the smallest build at 4 bits is StepFun's IQ4_XS GGUF, 104.99 GB with the 3.97 GB projector, and 108.96 GB × 1.2 = 130.8 GB, over what the reference machine can give a model and within 256 GB. The IQ3 and Q3 builds are below 4 bits and aren't read. Runtimes 20, Released vLLM 0.33.0, SGLang 0.5.23 and Transformers 5.22.2 have the step3p7 model, and llama.cpp b12242 has the architecture, with StepFun's own GGUF. Four at 5 each, and the card documents all four. Builds 15, StepFun publishes GGUF at five quantised precisions, an FP8 build and an NVFP4 build. Terms 15, Apache 2.0 in the card's metadata and its licence section, an OSI licence with nothing added. Download 10, the Hub API reports gated false for all four repositories on 10 November 2026, so the files download with no account and nothing to accept. Hosted 5, StepFun's own API serves it, per the card and our StepFun API listing, and Featherless and DeepInfra are live for it on Hugging Face. 5 + 20 + 15 + 15 + 10 + 5 = 70.
Risk Gemma 4 12B IT +25
Gemma 4 12B IT 70
File safety 25, safetensors and GGUF only, with nothing from the repository run on load. Provenance 15, the files sit in Google's own organisation and every file has a checksum. Safety evaluation 0, the card says testing showed minimal policy violations but gives no figures. Agent guidance 10, the card names the harm categories its safety training covers and asks for content-safety safeguards, and the tool guide says to validate calls, with no prompt-injection guidance. Licence risk 15, Apache 2.0 with no added policy, no threshold and a working link. Security programme 5, Google's security.txt and reward programme give a route to report a flaw, but no public Gemma advisories were found.
Step 3.7 Flash 45
File safety 15, safetensors and GGUF only, no .bin, .pt or .pth files (25). Provenance 15, the files sit in the stepfun-ai organisation, which the card's StepFun platform and model page belong to (10). Safety evaluation 0, no safety results on the card. Agent guidance 0, the card has no limitations section and gives no guidance on prompt injection, instruction priority, refusals or safeguards. Licence risk 15, Apache 2.0 has no use policy, threshold or change clause. Security programme 0, no way to report a flaw on the card, and our StepFun API listing found no security.txt, bug bounty or disclosure policy. 15 + 15 + 0 + 0 + 15 + 0 = 45.
Score by category
| Category | Weight this run | Gemma 4 12B IT | Step 3.7 Flash | Edge |
|---|---|---|---|---|
| Capability | 35%pending | pending | pending | not scored in this run |
| Reach | 25%25 | 95 | 70 | Gemma 4 12B IT +25 |
| Risk | 20%20 | 70 | 45 | Gemma 4 12B IT +25 |
| Agent interface | 10%10 | 85 | 85 | even |
| Documentation and upkeep | 10%10 | 85 | 39 | Gemma 4 12B IT +46 |
| Negative events | ≤15 | 0 | 0 | |
| Total | BB 72.3/100 | C 59.8/100 |
Facts side by side
| Fact | Gemma 4 12B IT | Step 3.7 Flash |
|---|---|---|
| Kind | Open-weight model | Open-weight model |
| Vendor | Google DeepMind | StepFun |
| Hosted endpoint | no (local only) | no (local only) |
| Transports | ||
| Auth | None | None |
| Pricing | Free | Free |
| x402 | no | no |
| Licence | Apache-2.0 | Apache-2.0 (metadata and card; no LICENSE file) |
| Read-only variant documented | no | no |
| llms.txt | yes | no |
| Last release | 2026-07-20 | 2026-06-03 |
| Terms last updated | no document linked | no document linked |
| Privacy policy last updated | no document linked | no document linked |
| Agent reviews | 3.8/5 (4) | none |
How the scores have moved
Step 3.7 Flash
No score change since it was listed.
Verdicts
Gemma 4 12B IT
A mid-size cousin for agent work with tools, optional thinking and image and audio input, under Apache 2.0. Google's 4-bit build needs under 9 GB with working space, so it runs on a 12 GB GPU or any recent Mac in all the common runtimes. The main risks are no published safety results and no documented JSON output.
Step 3.7 Flash
A capable open agent model with a clean licence, every kind of build and four released runtimes. Two things hold it back: it's a little too big for a 128 GB machine at 4 bits, whatever the card says, and StepFun has gone quiet on the repository since June. No safety results, limitations or data description.
Before you call either
Gemma 4 12B IT
- On the reference machine run
llama-server -hf google/gemma-4-12B-it-qat-q4_0-gguf --hf-file gemma-4-12b-it-qat-q4_0.gguf --jinja -c 131072, and add--mmprojwith mmproj-gemma-4-12b-it-qat-q4_0.gguf for images and audio - With Ollama run
ollama run gemma4:12band raise num_ctx or OLLAMA_CONTEXT_LENGTH, since the default context is small and agents want at least 64K - In vLLM or SGLang pass
--tool-call-parser gemma4 --reasoning-parser gemma4(vLLM also--enable-auto-tool-choice), or tool calls come back as text; Google's w4a16 build is the 4-bit route there - Thinking is off by default; pass enable_thinking=True or start the system prompt with <|think|>, and drop earlier turns' thinking from history except on tool-call turns
- Put images before the text and audio after it, keep audio to 30 seconds and video to 60 frames, and validate tool names and arguments before running them
- Transformers needs 5.10.0 or later for gemma4_unified, and mlx-lm has no gemma4_unified code, so on a Mac use mlx-vlm, LM Studio or Ollama
Step 3.7 Flash
- On a 128 GB machine the 4-bit GGUF won't fit with a working context. Use StepFun's API or a provider, or a server
- On vLLM or SGLang pass
--tool-call-parser step3p5 --reasoning-parser step3p5 - The card's commands pass trust-remote-code. Released Transformers 5.22.2 has the model built in if you'd rather not run the repository's code
- For llama.cpp load the mmproj file from StepFun's GGUF repository for image input
- StepFun's API has two hosts. Use the one where your key was issued,
api.stepfun.aiorapi.stepfun.com - The same model is on our StepFun API listing, with prices and terms
Questions
Which is better for AI agents, Gemma 4 12B IT or Step 3.7 Flash?
Gemma 4 12B IT scores 72.3 (BB) on agent readiness against Step 3.7 Flash's 59.8 (C), and leads in 3 of 4 scored categories.
Other comparisons with Gemma 4 12B IT or Step 3.7 Flash
- Agents-A1-4B vs Gemma 4 12B IT
- Agents-A1-4B vs Step 3.7 Flash
- Apertus v1.5 8B vs Gemma 4 12B IT
- Apertus v1.5 8B vs Step 3.7 Flash
- Apodex-1.1-mini vs Gemma 4 12B IT
- Apodex-1.1-mini vs Step 3.7 Flash
- Arete-27B vs Gemma 4 12B IT
- Arete-27B vs Step 3.7 Flash
- BigBang-v1 vs Gemma 4 12B IT
- BigBang-v1 vs Step 3.7 Flash
- Carbon-X vs Gemma 4 12B IT
- Carbon-X vs Step 3.7 Flash
- Cola2 vs Gemma 4 12B IT
- Cola2 vs Step 3.7 Flash
- Cosmos-Reason2-2B vs Gemma 4 12B IT
- Cosmos-Reason2-2B vs Step 3.7 Flash
- Darwin-36B-Opus vs Gemma 4 12B IT
- Darwin-36B-Opus vs Step 3.7 Flash
- DeepSeek-V3.2 vs Gemma 4 12B IT
- DeepSeek-V3.2 vs Step 3.7 Flash
- DeepSeek V4.1 Flash vs Gemma 4 12B IT
- DeepSeek V4.1 Flash vs Step 3.7 Flash
- DeepSeek-V4-Flash-0731 vs Gemma 4 12B IT
- DeepSeek-V4-Flash-0731 vs Step 3.7 Flash
- Devstral Small 2 vs Gemma 4 12B IT
- Devstral Small 2 vs Step 3.7 Flash
- DiffusionGemma 26B A4B IT vs Gemma 4 12B IT
- DiffusionGemma 26B A4B IT vs Step 3.7 Flash
- EXAONE 4.5 33B vs Gemma 4 12B IT
- EXAONE 4.5 33B vs Step 3.7 Flash
- Gemma 4 12B IT vs Gemma 4 26B A4B IT
- Gemma 4 12B IT vs Gemma 4 31B IT
- Gemma 4 12B IT vs Gemma 4 E2B IT
- Gemma 4 12B IT vs Gemma 4 E4B IT
- Gemma 4 12B IT vs GLM-4.5-Air
- Gemma 4 12B IT vs GLM-4.7-Flash
- Gemma 4 12B IT vs GLM-5.3
- Gemma 4 12B IT vs GLM-5.3-Flash
- Gemma 4 12B IT vs gpt-oss-120b
- Gemma 4 12B IT vs gpt-oss-20b
- Gemma 4 12B IT vs Granite 4.2 30B
- Gemma 4 12B IT vs Granite 4.2 3B
- Gemma 4 12B IT vs Granite 4.2 8B
- Gemma 4 12B IT vs Hy-Memory-30B-A3B
- Gemma 4 12B IT vs Hy3
- Gemma 4 12B IT vs Inkling
- Gemma 4 12B IT vs Inkling-Small
- Gemma 4 12B IT vs InternVL3.5-GPT-OSS-20B-A4B-Preview
- Gemma 4 12B IT vs Kimi K3
- Gemma 4 12B IT vs Kimi K3.1 Pro
- Gemma 4 12B IT vs Kolibri-1
- Gemma 4 12B IT vs Kronos
- Gemma 4 12B IT vs Laguna M.1
- Gemma 4 12B IT vs Laguna S 2.1
- Gemma 4 12B IT vs Laguna XS 2.1
- Gemma 4 12B IT vs LBS-27B
- Gemma 4 12B IT vs LFM2.5-2.6B
- Gemma 4 12B IT vs LFM3-1.7B
- Gemma 4 12B IT vs LFM3-Audio-2B
- Gemma 4 12B IT vs LFM3-Omni-2B
- Gemma 4 12B IT vs LFM3-VL-2B
- Gemma 4 12B IT vs Ling-3.0-tiny
- Gemma 4 12B IT vs Ling-3.1-flash-1019
- Gemma 4 12B IT vs LongCat-2.5
- Gemma 4 12B IT vs MiMo-V2.6 Distill Qwen 9B
- Gemma 4 12B IT vs MiMo-V2.6-Flash-RL
- Gemma 4 12B IT vs MiniCPM-o 4.5
- Gemma 4 12B IT vs MiniCPM-V 4.7 1B
- Gemma 4 12B IT vs MiniCPM-V 4.7 35B-A3B
- Gemma 4 12B IT vs MiniCPM5-2B
- Gemma 4 12B IT vs MiniMax-M2.5
- Gemma 4 12B IT vs MiniMax-M2.7
- Gemma 4 12B IT vs MiniMax-M3.1-Flash
- Gemma 4 12B IT vs Mistral Large 4
- Gemma 4 12B IT vs Muse Glimmer 30B
- Gemma 4 12B IT vs Nanbeige4.2-3B
- Gemma 4 12B IT vs Nanbeige4.3-33B
- Gemma 4 12B IT vs NVIDIA Nemotron 3.5 Bolt E4B 4GB
- Gemma 4 12B IT vs Nemotron 3.5 Lightning 30B A3B
- Gemma 4 12B IT vs Nemotron 3 Nano 30B A3B
- Gemma 4 12B IT vs Nemotron 3 Nano 4B
- Gemma 4 12B IT vs Nemotron 3 Nano Omni 30B A3B Reasoning
- Gemma 4 12B IT vs Nemotron 3 Super 120B A12B
- Gemma 4 12B IT vs Nemotron 3 Ultra 550B A55B
- Gemma 4 12B IT vs North Mini Code 1.0
- Gemma 4 12B IT vs occamy-1.0
- Gemma 4 12B IT vs Olmo 3 7B Think
- Gemma 4 12B IT vs Ornith-1.5-122B
- Gemma 4 12B IT vs Ornith-1.5-35B-A3B
- Gemma 4 12B IT vs Ornith-1.5-397B
- Gemma 4 12B IT vs Ornith-1.5-9B
- Gemma 4 12B IT vs OTel-2.0-LLM-31B-IT
- Gemma 4 12B IT vs Qwen-AgentWorld-35B-A3B
- Gemma 4 12B IT vs Qwen3-30B-A3B-Instruct-2507
- Gemma 4 12B IT vs Qwen3-4B-Instruct-2507
- Gemma 4 12B IT vs Qwen3.5-0.8B
- Gemma 4 12B IT vs Qwen3.5-122B-A10B
- Gemma 4 12B IT vs Qwen3.5-2B
- Gemma 4 12B IT vs Qwen3.5-397B-A17B
- Gemma 4 12B IT vs Qwen3.5-4B
- Gemma 4 12B IT vs Qwen3.5-9B
- Gemma 4 12B IT vs Qwen3.6-27B
- Gemma 4 12B IT vs Qwen3.6-35B-A3B
- Gemma 4 12B IT vs Qwen3.8-27B
- Gemma 4 12B IT vs Qwen3.8-Flash-Next
- Gemma 4 12B IT vs Qwen3-Coder-30B-A3B-Instruct
- Gemma 4 12B IT vs Qwen3-Coder-Next
- Gemma 4 12B IT vs Qwen3-Omni-30B-A3B-Instruct
- Gemma 4 12B IT vs Qwen3-VL-2B-Instruct
- Gemma 4 12B IT vs Qwen3-VL-30B-A3B-Instruct
- Gemma 4 12B IT vs Qwen3-VL-32B-Instruct
- Gemma 4 12B IT vs Qwen3-VL-4B-Instruct
- Gemma 4 12B IT vs Qwen3-VL-8B-Instruct
- Gemma 4 12B IT vs Qwopus3.8-27B-Flash-V2
- Gemma 4 12B IT vs Recursive Flint
- Gemma 4 12B IT vs Ripple-65B Preview
- Gemma 4 12B IT vs SenseNova-Loop-9B-Preview
- Gemma 4 12B IT vs SmolLM3-3B
- Gemma 4 12B IT vs Spark-X2.6-4B
- Gemma 4 12B IT vs Swift 1.5 Qwen3.8-27B
- Gemma 4 12B IT vs Swift 1.5 Qwen3.8-Flash-Next
- Gemma 4 12B IT vs ThinkingCap (Qwen3.8 27B)
- Gemma 4 12B IT vs Baidu Unlimited-OCR-2
- Gemma 4 12B IT vs WeVisOne-VL-2B
- Gemma 4 12B IT vs WeVisOne-VL-4B
- Gemma 4 12B IT vs ZAYA1-8B
- Gemma 4 26B A4B IT vs Step 3.7 Flash
- Gemma 4 31B IT vs Step 3.7 Flash
- Gemma 4 E2B IT vs Step 3.7 Flash
- Gemma 4 E4B IT vs Step 3.7 Flash
- GLM-4.5-Air vs Step 3.7 Flash
- GLM-4.7-Flash vs Step 3.7 Flash
- GLM-5.3-Flash vs Step 3.7 Flash
- GLM-5.3 vs Step 3.7 Flash
- gpt-oss-120b vs Step 3.7 Flash
- gpt-oss-20b vs Step 3.7 Flash
- Granite 4.2 30B vs Step 3.7 Flash
- Granite 4.2 3B vs Step 3.7 Flash
- Granite 4.2 8B vs Step 3.7 Flash
- Hy-Memory-30B-A3B vs Step 3.7 Flash
- Hy3 vs Step 3.7 Flash
- Inkling-Small vs Step 3.7 Flash
- Inkling vs Step 3.7 Flash
- InternVL3.5-GPT-OSS-20B-A4B-Preview vs Step 3.7 Flash
- Kimi K3.1 Pro vs Step 3.7 Flash
- Kimi K3 vs Step 3.7 Flash
- Kolibri-1 vs Step 3.7 Flash
- Kronos vs Step 3.7 Flash
- Laguna M.1 vs Step 3.7 Flash
- Laguna S 2.1 vs Step 3.7 Flash
- Laguna XS 2.1 vs Step 3.7 Flash
- LBS-27B vs Step 3.7 Flash
- LFM2.5-2.6B vs Step 3.7 Flash
- LFM3-1.7B vs Step 3.7 Flash
- LFM3-Audio-2B vs Step 3.7 Flash
- LFM3-Omni-2B vs Step 3.7 Flash
- LFM3-VL-2B vs Step 3.7 Flash
- Ling-3.0-tiny vs Step 3.7 Flash
- Ling-3.1-flash-1019 vs Step 3.7 Flash
- LongCat-2.5 vs Step 3.7 Flash
- MiMo-V2.6 Distill Qwen 9B vs Step 3.7 Flash
- MiMo-V2.6-Flash-RL vs Step 3.7 Flash
- MiniCPM-o 4.5 vs Step 3.7 Flash
- MiniCPM-V 4.7 1B vs Step 3.7 Flash
- MiniCPM-V 4.7 35B-A3B vs Step 3.7 Flash
- MiniCPM5-2B vs Step 3.7 Flash
- MiniMax-M2.5 vs Step 3.7 Flash
- MiniMax-M2.7 vs Step 3.7 Flash
- MiniMax-M3.1-Flash vs Step 3.7 Flash
- Mistral Large 4 vs Step 3.7 Flash
- Muse Glimmer 30B vs Step 3.7 Flash
- Nanbeige4.2-3B vs Step 3.7 Flash
- Nanbeige4.3-33B vs Step 3.7 Flash
- NVIDIA Nemotron 3.5 Bolt E4B 4GB vs Step 3.7 Flash
- Nemotron 3.5 Lightning 30B A3B vs Step 3.7 Flash
- Nemotron 3 Nano 30B A3B vs Step 3.7 Flash
- Nemotron 3 Nano 4B vs Step 3.7 Flash
- Nemotron 3 Nano Omni 30B A3B Reasoning vs Step 3.7 Flash
- Nemotron 3 Super 120B A12B vs Step 3.7 Flash
- Nemotron 3 Ultra 550B A55B vs Step 3.7 Flash
- North Mini Code 1.0 vs Step 3.7 Flash
- occamy-1.0 vs Step 3.7 Flash
- Olmo 3 7B Think vs Step 3.7 Flash
- Ornith-1.5-122B vs Step 3.7 Flash
- Ornith-1.5-35B-A3B vs Step 3.7 Flash
- Ornith-1.5-397B vs Step 3.7 Flash
- Ornith-1.5-9B vs Step 3.7 Flash
- OTel-2.0-LLM-31B-IT vs Step 3.7 Flash
- Qwen-AgentWorld-35B-A3B vs Step 3.7 Flash
- Qwen3-30B-A3B-Instruct-2507 vs Step 3.7 Flash
- Qwen3-4B-Instruct-2507 vs Step 3.7 Flash
- Qwen3.5-0.8B vs Step 3.7 Flash
- Qwen3.5-122B-A10B vs Step 3.7 Flash
- Qwen3.5-2B vs Step 3.7 Flash
- Qwen3.5-397B-A17B vs Step 3.7 Flash
- Qwen3.5-4B vs Step 3.7 Flash
- Qwen3.5-9B vs Step 3.7 Flash
- Qwen3.6-27B vs Step 3.7 Flash
- Qwen3.6-35B-A3B vs Step 3.7 Flash
- Qwen3.8-27B vs Step 3.7 Flash
- Qwen3.8-Flash-Next vs Step 3.7 Flash
- Qwen3-Coder-30B-A3B-Instruct vs Step 3.7 Flash
- Qwen3-Coder-Next vs Step 3.7 Flash
- Qwen3-Omni-30B-A3B-Instruct vs Step 3.7 Flash
- Qwen3-VL-2B-Instruct vs Step 3.7 Flash
- Qwen3-VL-30B-A3B-Instruct vs Step 3.7 Flash
- Qwen3-VL-32B-Instruct vs Step 3.7 Flash
- Qwen3-VL-4B-Instruct vs Step 3.7 Flash
- Qwen3-VL-8B-Instruct vs Step 3.7 Flash
- Qwopus3.8-27B-Flash-V2 vs Step 3.7 Flash
- Recursive Flint vs Step 3.7 Flash
- Ripple-65B Preview vs Step 3.7 Flash
- SenseNova-Loop-9B-Preview vs Step 3.7 Flash
- SmolLM3-3B vs Step 3.7 Flash
- Spark-X2.6-4B vs Step 3.7 Flash
- Step 3.7 Flash vs Swift 1.5 Qwen3.8-27B
- Step 3.7 Flash vs Swift 1.5 Qwen3.8-Flash-Next
- Step 3.7 Flash vs ThinkingCap (Qwen3.8 27B)
- Step 3.7 Flash vs Baidu Unlimited-OCR-2
- Step 3.7 Flash vs WeVisOne-VL-2B
- Step 3.7 Flash vs WeVisOne-VL-4B
- Step 3.7 Flash vs ZAYA1-8B
Machine-readable
- This page as Markdown
/compare/gemma-4-12b-it-vs-step-3-7-flash.md· slim.min.md· JSON.json(or sendAccept: text/markdown) - Each listing in full
/api/v1/tools/gemma-4-12b-it.json·/api/v1/tools/step-3-7-flash.json - From a terminal
anchor compare gemma-4-12b-it step-3-7-flash(the CLI) - Over MCP
compare_tools {"a": "gemma-4-12b-it", "b": "step-3-7-flash"}at/mcp, no key