Gemma 4 31B on Cerebras: 1,800+ TPS — The Fastest Multimodal Inference Yet

Update — July 16, 2026: Gemma 4 31B agentic benchmarks improved — +10.1% Tau2 Telecom, +4.5% TB2 Agents, plus max_soft_tokens 1120 for sharper OCR on refreshed HF weights.

June 29, 2026: Cerebras announced Gemma 4 31B running at 1,851 output tokens per second on Cerebras Inference — 35× a typical GPU endpoint per Artificial Analysis benchmarking. It is the first Google DeepMind model on the platform and the first multimodal model at wafer-scale speed: developers can feed images — screenshots, documents, charts, UI states — into inference that previously only text models achieved at this throughput.

For teams building visual agent loops while Fable 5 remains offline and GLM-5.3 vision is still a community wishlist, Gemma 4 on Cerebras is a different bet: open-weight multimodal at real-time latency, not frontier closed API access.

TL;DR — Gemma 4 on Cerebras

Why Cerebras Paired With Gemma 4

Cerebras has benchmarked fast inference across open-weight stacks — Kimi, GLM, GPT-OSS, Qwen — on wafer-scale hardware. Gemma 4 is the first Google DeepMind model on the platform, and the first where vision enters the loop at Cerebras speed.

Olivier Lacombe, Product Lead for Gemma at Google DeepMind:

"Gemma 4, Google DeepMind's family of open models, was built to bring advanced reasoning and multimodal capabilities at developer-friendly sizes. Pairing these capabilities with Cerebras's wafer-scale technology provides developers with an exciting platform for running extremely fast visual and agentic workflows."

The pitch is not "same product, faster." Logan Kilpatrick (Google DeepMind):

"If every model was doing 2,000 tokens per second, you would probably build different products. You wouldn't build the same product and just have it be faster."

At 1,800+ TPS, multimodal agent loops — inspect image → reason → structured output → tool call → verify → retry — stop feeling like batch jobs and start feeling interactive.

Speed Numbers — What 1,851 TPS Changes

Cerebras recommends Gemma 4 31B as the reference medium-size model on its cloud: an alternative to Haiku, GPT-OSS, or Llama with equal or higher intelligence at Cerebras speed.

Dense vs MoE: Gemma 4 31B is a dense multimodal model — high intelligence without the large memory footprint of MoE serving. That fits Cerebras's wafer-scale serving story: strong enough for serious agent work, efficient to run at scale, Apache 2.0 for build-around freedom.

Multimodal on Wafer-Scale — A Platform First

Before Gemma 4, Cerebras Inference was text-first at extreme speed. Gemma 4 adds:

• Screenshot understanding — dashboards, dense UI, document pages
• Chart and diagram reasoning
• Scanned pages and forms
• Computer use and robotics pathways (per Cerebras)

Cerebras states multimodal support starts with Gemma 4 and will extend to additional models — the same platform pattern they used for Kimi and GLM text inference.

This directly addresses the workflow gap Zhipu users asked for in GLM-5.3 polls: vision integrated with reasoning, not a separate VL model bridge. Gemma 4 on Cerebras ships that integration today — on cloud wafer hardware, not local 16GB VRAM.

For on-device multimodal, see Gemma 4 12B local guide and Gemma 4 E4B + Argent iOS automation.

Example Workloads Cerebras Highlights

Screenshot to Insight

Feed a dense dashboard screenshot or document page. The model identifies what matters, explains the finding, returns structured output — in real time, not after a GPU wait.

Long-context summarization

Hand it a research report or technical brief. Get a decision-ready summary fast enough to read, react, and re-query in one sitting — relevant for teams comparing against Kimi K2.7's long-context coding or GLM document pipelines.

Screenshot to Patch

Play to medium-model strengths: broken UI screenshot + source + console error → minimal patch + verification checks. This is the agentic coding loop Fable 5 marketed — now runnable on open weights at speeds that keep a human in the loop.

Computer use and robotics

Gemma 4's multimodal stack supports UI state reasoning — overlapping with E4B + Argent simulator navigation at the edge, but at cloud scale and speed for heavier agents.

Why Agent Loops Compound at 1,800 TPS

Multimodal and agentic workflows rarely call a model once:

Visual input → reason → structured output → tool call → check result → retry

At conventional GPU speeds, each loop adds seconds — too slow for real-time copilots, live UI repair, or tight human-in-the-loop review.

At 1,800+ TPS:

• Front-end iteration feels near-instant
• Document and screenshot workflows return in a fraction of prior latency
• More verification steps and retries fit the same product budget

The product implication is architectural: you design for continuous visual reasoning, not "send screenshot, wait, hope."

Gemma 4 31B vs Haiku 4.5 — The Open-Weight Angle

For international developers blocked from Fable 5, Gemma 4 + Cerebras is another unrestricted multimodal path — alongside GLM-5.2 text and Kimi K2.7-Code coding.

Haiku remains marginally smarter on the AA Index. Gemma 4 wins on speed, openness, and self-host/build-around freedom when paired with Cerebras cloud.

Cerebras Platform Context — Kimi, GLM, and the Speed Ladder

Cerebras has positioned itself as the inference speed leader for open weights:

• Kimi, GLM, GPT-OSS, Qwen — text at wafer-scale TPS
• Gemma 4 31B — first Google model, first multimodal

If your stack already routes GLM-5.2 through Z.ai for text coding, Gemma 4 on Cerebras is the vision complement — similar to the Qwen-VL → GLM bridge developers want eliminated in GLM-5.3, but with single-model multimodal at extreme speed.

Availability — Public Preview

Gemma 4 31B is on the Cerebras Inference Cloud in public preview for a limited time as of June 29, 2026.

Cerebras asks teams with workloads in:

• Multimodal reasoning
• Fast document processing
• Real-time audio and video (future platform extensions)

…to contact them directly via cerebras.ai.

This is not a self-serve unlimited free tier announcement — preview access with enterprise outreach for heavy multimodal pipelines.

What Developers Should Do

Need multimodal agents at interactive speed?

Evaluate Cerebras Inference Cloud preview for Gemma 4 31B. Benchmark your screenshot-to-patch or dashboard-insight loop against GPU baselines — Cerebras claims 35× on output TPS.

Need local / privacy-first multimodal?

Use Gemma 4 12B locally — 16GB VRAM, Apache 2.0, unified architecture. Trade wafer-scale TPS for data never leaving your machine.

Need frontier coding without vision?

GLM-5.2 and Kimi K2.7-Code remain the open-weight coding leaders while Fable is suspended.

Building computer-use agents?

Compare Gemma 4 on Cerebras (cloud speed) vs Gemma 4 E4B + Argent (local iOS simulator) — same family, different deployment surface.

The Honest Answer

Is Gemma 4 on Cerebras the fastest multimodal inference available?

Per Cerebras and Artificial Analysis methodology cited in the announcement — yes, as of June 29, 2026, at 1,851 output TPS for Gemma 4 31B.

Does it replace Fable 5 or Opus for coding?

No — it targets Haiku-class medium intelligence with multimodal speed. Serious autonomous coding at Fable depth still points to restricted US models or open alternatives (Kimi, GLM, LongCat-2.0).

Does it matter for product design?

Yes. Kilpatrick's quote is the thesis: at 2,000 TPS, you build different products — not the same agent with shorter waits. Multimodal loops become real-time collaborators.

Related Reading

Gemma ecosystem

• Gemma 4 July 2026 update — FA4, tool calling, vision
• Gemma 4 12B — Local Multimodal Guide
• Gemma 4 E4B + Argent iOS Automation
• Gemma 4 Open Duck Mini Robot
• DiffusionGemma 4× Faster Text Generation

Open-weight alternatives

• GLM-5.2 and the Fable 5 Ban
• GLM-5.3 Vision Community Poll
• Kimi K2.7-Code

Agent context

• Is Fable 5 Back?
• Claude Code vs Codex vs Gemini CLI vs GLM-5.2

Speed and intelligence figures cite Cerebras's June 29, 2026 announcement and Artificial Analysis benchmarking as referenced by Cerebras. Preview availability and pricing may change — verify on cerebras.ai before production commitments.