Gemma 4 31B
Gemma 4 31B needs 20.9 GB of VRAM at Q4_K_M with 8k context, computed from its config.json as checked on 3 November 2026. The smallest card that holds it has 24 GB, a tight fit, and the fastest desktop card that runs it, the RTX 5090, generates about 69 tokens per second.
- Parameters
- 31.3B
- Context
- 262,144 tokens
- Licence
- Apache 2.0
- Released
- March 2026
- Config checked
- 3 November 2026
Open Gemma 4 31B in the LLM checker
Which card size runs it
20.9 GB against each memory size. A model may use 95% of a card, and only what is left over, or missing, is coloured; tight means above 85%, with little room for a longer conversation.
Other quantisations, context lengths or a Q8 KV cache: the VRAM calculator.
Computed 3 November 2026Which GPU to pick
The smallest card here that loads it at Q4_K_M with 8k of context is the RTX 5090 Laptop GPU (24 GB): 20.9 GB needed, an estimated 35 tokens per second, and a tight fit that leaves little for the conversation. For room to work, meaning 32,768 tokens of context while staying under 85% of memory, step up to the RTX 5090 (32 GB), which needs 22.9 GB for that and generates at about 69 tokens per second.
At Q8_0, which is near-lossless, the card is the RTX A6000 (48 GB), needing 36.5 GB. On Apple silicon, a Mac mini M4 Pro holds it with 32,768 tokens of context at an estimated 11 tokens per second; prompt processing is slower than on Nvidia, which long documents make obvious. Among unified-memory desktops, the Ryzen AI Max+ 395 holds it at 32,768 tokens in 22.9 GB, at an estimated 10 tokens per second.
Speed on each card
| GPU | Fits | Est. tokens/s | Feels like | Memory | Bandwidth |
|---|---|---|---|---|---|
| H100 SXM | Yes | ≤ 129 | faster than you can read | 80 GB | 3,350 GB/s |
| A100 | Yes | 79 | faster than you can read | 80 GB | 2,039 GB/s |
| RTX 5090 | Yes | 69 | faster than you can read | 32 GB | 1,792 GB/s |
| RTX PRO 6000 Blackwell | Yes | 69 | faster than you can read | 96 GB | 1,792 GB/s |
| RTX PRO 5000 Blackwell | Yes | 52 | comfortable for chat | 48 GB | 1,344 GB/s |
| RTX 4090 | Tight | 39 | comfortable for chat | 24 GB | 1,008 GB/s |
| RX 7900 XTX | Tight | 37 | comfortable for chat | 24 GB | 960 GB/s |
| RTX 6000 Ada | Yes | 37 | comfortable for chat | 48 GB | 960 GB/s |
| RTX 3090 | Tight | 36 | comfortable for chat | 24 GB | 936 GB/s |
| RTX 5090 Laptop GPU | Tight | 35 | comfortable for chat | 24 GB | 896 GB/s |
| RTX PRO 4500 Blackwell | Yes | 35 | comfortable for chat | 32 GB | 896 GB/s |
| L40S | Yes | 33 | comfortable for chat | 48 GB | 864 GB/s |
| RTX A6000 | Yes | 30 | comfortable for chat | 48 GB | 768 GB/s |
| RTX PRO 4000 Blackwell | Tight | 26 | comfortable for chat | 24 GB | 672 GB/s |
| Radeon AI PRO R9700 | Yes | 25 | usable, a little slow | 32 GB | 640 GB/s |
| Arc Pro B70 | Yes | 23 | usable, a little slow | 32 GB | 608 GB/s |
| Arc Pro B60 | Tight | 18 | usable, a little slow | 24 GB | 456 GB/s |
| RTX PRO 4000 Blackwell SFF | Tight | 17 | usable, a little slow | 24 GB | 432 GB/s |
| RTX Spark | Yes | 12 | slow; fine for batch jobs | 128 GB (102 for the GPU) | 300 GB/s |
| DGX Spark | Yes | 11 | slow; fine for batch jobs | 128 GB (126 for the GPU) | 273 GB/s |
| Ryzen AI Max+ 395 | Yes | 10 | slow; fine for batch jobs | 128 GB (96 for the GPU) | 256 GB/s |
| RX 7900 XT | No | - | does not fit | 20 GB | 800 GB/s |
| RTX 4060 Ti | No | - | does not fit | 16 GB | 288 GB/s |
| RTX 5060 Ti | No | - | does not fit | 16 GB | 448 GB/s |
| RTX 5080 | No | - | does not fit | 16 GB | 960 GB/s |
| RTX 5070 Ti | No | - | does not fit | 16 GB | 896 GB/s |
| RTX 4080 SUPER | No | - | does not fit | 16 GB | 736 GB/s |
| RX 9070 XT | No | - | does not fit | 16 GB | 640 GB/s |
| RX 9060 XT | No | - | does not fit | 16 GB | 320 GB/s |
| RTX 3060 | No | - | does not fit | 12 GB | 360 GB/s |
| RTX 4070 | No | - | does not fit | 12 GB | 504 GB/s |
| RTX 5070 | No | - | does not fit | 12 GB | 672 GB/s |
| Arc B580 | No | - | does not fit | 12 GB | 456 GB/s |
Single-stream decode, from memory bandwidth at 70% efficiency. Prompt processing and batching not included. These are ceilings, not measurements: Gemma 4 31B reads only 18.2 GB of weights per token at Q4, so little that the per-token costs the formula leaves out decide much of the real speed. The figure marked ≤ is where a real runtime falls furthest below the number shown. See the speed estimator for other quantisations and Apple hardware, or every card compared if you are choosing hardware rather than a model.
Computed 3 November 2026Memory by quantisation and context
| Quant | 4,096 ctx | 8,192 ctx | 32,768 ctx |
|---|---|---|---|
| Q8_0 | 36.2 GB · fits 48 GB | 36.5 GB · fits 48 GB | 38.5 GB · fits 48 GB |
| Q5_K_M | 24.8 GB · fits 32 GB | 25.1 GB · fits 32 GB | 27.1 GB · fits 32 GB |
| Q4_K_M | 20.6 GB · fits 24 GB | 20.9 GB · fits 24 GB | 22.9 GB · fits 32 GB |
Weights at this quant: 18.2 GB at Q4. Every extra 1,000 tokens of context adds about 0.082 GB of KV cache at FP16, on top of a fixed 0.839 GB that the sliding-window layers hold once their window is full. Try other settings in the VRAM calculator.
About the model
Gemma 4 31B has 31.3B parameters. It has 60 layers: 50 use sliding-window attention over the last 1,024 tokens (16 KV heads of dimension 256), and 10 are global layers with 4 KV heads of dimension 512. The dense flagship of Google DeepMind's Gemma 4 family, spring 2026. Vision-language, a 256k window, configurable thinking, and an Apache licence in place of Gemma 3's custom terms.
Text and images 262,144-token window Apache 2.0 March 2026
Reasoning. Configurable thinking modes, per the model card.
A runtime without a sliding-window cache stores every layer at full length and needs far more than this at long context. The parameter count includes the vision encoder; a text-only GGUF is slightly smaller.
Native context window: 262,144 tokens. Licence: Apache 2.0, a permissive licence. Weights published in March 2026. It follows Gemma 3 27B.
Good at
- The hardest work a 24 GB card can take on: a dense 31B that still fits one at Q4.
- Long context at moderate cost: 50 of its 60 layers are sliding-window.
- Commercial use under Apache 2.0.
Watch out for
- It fills a 24 GB card. It loads and runs, but a 32 GB card is where it has room for a long document.
- Its ten global layers keep a full-length cache (4 KV heads of dimension 512), so per token it costs more context memory than Qwen3.8 27B does.
- No audio input; in Gemma 4 that is the 12B.
Similar models
The models that need about the same memory as Gemma 4 31B at Q4_K_M and 8k, which makes them the real alternatives on whatever card you have. The context column is what each extra 1,000 tokens costs, and it is where models of the same size differ most.
| Model | Size | Needs | Per 1k context | Window | Released |
|---|---|---|---|---|---|
| Gemma 4 31B | 31.3B | 20.9 GB | 0.082 GB | 256k | 2026-03 |
| Granite 4.2 30B | 29.3B | 20.3 GB | 0.262 GB | 128k | 2026-08 |
| MiniCPM-V 4.7 35B-A3B (MoE) | 35.2B · 3B active | 22 GB | 0.02 GB | 256k | 2026-10 |
| GLM-4.7 Flash 30B-A3B (MoE) | 31.2B · 3B active | 19.8 GB | 0.054 GB | 198k | 2026-01 |
| Qwen3 30B-A3B (MoE) | 30.5B · 3.3B active | 19.7 GB | 0.098 GB | 40k | 2025-04 |
See every model compared, which models fit on 8 to 96 GB, or the cheapest hour of every cloud card.
Questions
- Can I run Gemma 4 31B on a 24 GB card?
- Yes, at Q4_K_M and 8k context it needs about 20.9 GB, which fits a 24 GB card tightly.
- How much VRAM does Gemma 4 31B need at Q8?
- About 36.5 GB at 8k context, or 38.5 GB at 32k. Q8 is near-lossless; use it when it fits.
- What is the context length of Gemma 4 31B?
- 262,144 tokens natively. Holding all of it at Q4_K_M takes about 41.7 GB, of which 22.3 GB is context.
- Why does Gemma 4 31B need so little memory for long context?
- Because most of its layers do not keep a cache that grows. It has 60 layers: 50 use sliding-window attention over the last 1,024 tokens (16 KV heads of dimension 256), and 10 are global layers with 4 KV heads of dimension 512. At 32k tokens the context costs about 3.5 GB; if all 60 layers kept a conventional full-length cache, the same conversation would cost about 32.2 GB. That saving depends on the runtime implementing the sliding-window cache. llama.cpp and vLLM do; a build that does not will use far more.
- How fast is Gemma 4 31B on an RTX 4090?
- At most about 39 tokens per second at Q4, single stream, which is comfortable for chat. That is a ceiling worked out from the card's memory bandwidth, not a measurement.
- Architecture values from config.json in google/gemma-4-31B-it on Hugging Face, checked 3 November 2026. Every figure on this page is computed from those values; check them before buying hardware for this model.
- Estimates assume a single conversation on a card that is otherwise free. A desktop on the same GPU takes 0.5 to 2 GB.