Can you run Gemma 4 26B-A4B on RTX 4090?

Gemma 4 26B-A4B Q4_K_M on the NVIDIA GeForce RTX 4090: verdict, VRAM math and estimated speed.

Yes, it runs
Quick answer

Yes, the NVIDIA GeForce RTX 4090 runs Gemma 4 26B-A4B. 16 GB weights at Q4_K_M vs 21.6 GB usable VRAM; ~85 tok/s est. (ModelFit, 2026).

$ollama run gemma4:26b
VERDICT
Fits
EST. SPEED
~85 tok/s
WEIGHTS
16 GB Q4_K_M

VRAM math and speed are ModelFit engine estimates, not measurements. Commands are registry-verified Ollama tags.

Cite this page: ModelFit, Gemma 4 26B-A4B on RTX 4090, https://modelfit.io/can-i-run/gemma4-26b-a4b-q4-on-rtx-4090/, updated September 2026, CC BY 4.0.

Last updated: September 18, 2026 · Editor: ModelFit Team

VRAM
24 GB (21.6 usable)
Model weights
16 GB Q4_K_M
Est. speed
~85 tok/s
First token
~0.4s
Fit grade
B · 26

What limits this combo

The four constraints the fit engine checks for Gemma 4 26B-A4B on the RTX 4090, in order of what usually breaks first.

Weights fitOK16 GB of weights against 21.6 GB usable — 5.6 GB of headroom for context and the runtime.
Context ceilingTightContext fits up to 16k tokens; the next tier tips the total over 21.6 GB. Long documents and RAG prompts are what hit this wall first.
SpeedOK~85 tok/s (est.) on 1008 GB/s of memory bandwidth — comfortable for interactive chat and agentic loops.
Use-case ceilingTightStrongest fit: Chat (A). Agentic coding caps at C — either the speed floor or the model's tuning is the limit; see the workload table for the per-case reason.

Quant explorer

Switch between the quality-gated builds tracked for this exact combo. Q4_K_M stays the default recommendation; heavier Q6/Q8 builds only appear when their exact Ollama tags are registry-verified.

ModelFit engine estimate
Verdict
Runs
Est. speed
~85 tok/s
First token
~0.4s · Instant
Safe context
16k
B · 26 Comfortable fit

Why no 2-bit builds: Q1/Q2-class quants can look attractive in a memory table, but their quality loss is large enough that ModelFit excludes them from rankings instead of inflating the catalog with junk options. Usable VRAM: 21.6 GB.

Workload verdicts

Gemma 4 26B-A4B on the RTX 4090, graded per use case from the model's registry-verified tuning and the engine's speed estimate for this exact combo.

ModelFit engine estimate
ChatA~85 tok/s est. vs ~20 needed for chat.
CodingA~85 tok/s est. vs ~15 needed for coding.
Agentic codingC~85 tok/s est. vs ~30 needed for agentic coding; not tuned for tool-calling loops.
ReasoningC~85 tok/s est. vs ~12 needed for reasoning; not a reasoning-tuned model.
RAGC~85 tok/s est. vs ~20 needed for rag; 32k context does not fit.

Grades combine tag-verified tuning (a model not built for the workload caps at C) with the tok/s floor each workload needs to feel usable. A combo that partially offloads caps at C; one that does not fit is D everywhere.

Memory math: weights + context vs budget

Weights take 16 GB. Context costs extra KV-cache on top — this is where long-context sessions break on cards that technically fit the weights.

Weights 16 GBKV cache · 16k context 4.0 GBHeadroom 1.6 GBRuntime reserve 2.4 GB

The model leaves about 1.6 GB of the usable VRAM budget free after weights and 16k context.

ContextKV-cacheTotalFits
8k tokens2.0 GB18.0 GBFits
16k tokens4.0 GB20.0 GBFits
32k tokens8.0 GB24.0 GBOver
64k tokens16.0 GB32.0 GBOver
128k tokens32.0 GB48.0 GBOver

KV-cache figures assume an fp16 cache, the llama.cpp/Ollama default. Standard GQA models use a size-class estimate (8 KV heads x 128 head dim class); hybrid linear-attention models (Qwen3.5/3.6, Qwen3-Next) use the exact per-token cost from their published config, since only their sparse full-attention layers cache KV. A q8_0 KV cache roughly halves either figure. Estimates, not measurements.

See how fast it feels

A deterministic typing simulation for Gemma 4 26B-A4B: first token ~0.4s (instant prefill), then ~85 tokens/sec.

Start the simulation to preview the response pace.

ModelFit engine estimate, not a measured benchmark. Real speed varies with prompt length, thermals, runtime, and KV-cache settings.

Gemma 4 26B-A4B on RTX 4090: FAQ

Can the NVIDIA GeForce RTX 4090 run Gemma 4 26B-A4B?

Yes. Gemma 4 26B-A4B (Q4_K_M) loads in about 16 GB and the RTX 4090 offers 21.6 GB of usable VRAM, leaving headroom for context. Expect at roughly 85 tokens/sec (est.).

How much VRAM does Gemma 4 26B-A4B need?

About 16 GB for the weights at Q4_K_M, plus KV-cache for context: roughly 4.0 GB extra at 16k tokens. The RTX 4090 budget is 21.6 GB (24 GB x 90%).

What is the best quantization of Gemma 4 26B-A4B for the RTX 4090?

Stick with the Q4_K_M build at 16 GB — every heavier quant exceeds the 21.6 GB usable VRAM.

What GPU do I need to run Gemma 4 26B-A4B comfortably?

The RTX 4090 already runs Gemma 4 26B-A4B comfortably. Larger cards only buy you longer context or a heavier quant.

How much context can Gemma 4 26B-A4B use on the RTX 4090?

Up to 16k tokens stay fully in VRAM (20.0 GB total with the KV cache). At 32k tokens the total reaches 24.0 GB and tips over the budget. Quantizing the KV cache to q8_0 roughly halves the KV column and buys back about one context tier.

Is Gemma 4 26B-A4B on the RTX 4090 fast enough for coding agents?

Barely on paper: the engine estimates ~85 tok/s against the ~30 tok/s an agentic loop needs to feel responsive, but Gemma 4 26B-A4B is not tuned for tool-calling, which caps its agentic grade — a coding-tuned model is the better pick here.