Google
ChatCodingReasoningMulti-purpose25.2B
Chat

Gemma 4 26B-A4B

Gemma · Apache 2.0

Mixture-of-Experts architecture: 25.2B total parameters but only 3.8B active per token (8 selected + 1 shared expert per layer, out of 128 total). Hybrid dense+sparse FFN design. Inference throughput closer to a 4B dense model; quality closer to a 27B dense model. 256K context window. Benchmarks: 88.3% AIME 2026, 82.6% MMLU Pro, 77.1% LiveCodeBench. All 25.2B weights must be loaded into VRAM despite sparse activation; fits on 24 GB GPUs at Q4_K_M. Apache 2.0 licensed.

Parameters
25.2B
Architecture
Dense
Context
256,000 tokens
Released
2026-04-02
Engines
llama.cpp, ollama, vLLM
Builder Tools
Claude Code, Codex CLI, Continue, Cursor, LM Studio, Ollama, Open WebUI, Windsurf

Parameters

25.2B

VRAM

24 GB

Context

250K

Formats

5

GPUs

37

Gemma 4 26B-A4B (25.2B) requires 24 GB VRAM at recommended quality (Q6_K). At efficient quality (Q4_K_M), it fits in 18 GB VRAM, making it compatible with the NVIDIA RTX 4090 Laptop (150-175W). On NVIDIA Grace Blackwell Ultra GB300, expect approximately 500 tok/s at Q8_0. For the best experience, AMD AI Powerhouse ($1,818) is recommended.

Source: OwnRig methodology

VRAM (Recommended)

24 GB

Quantization

Q6_K

File Size

22.86 GB

Max Context

250K tokens

Primary Use

Chat

Memory

VRAM Requirements

QualityQuantizationVRAMFile Size
fullQ8_028 GB26.86 GB
recommendedQ6_K24 GB22.86 GB
recommendedQ5_K_M20.5 GB19.32 GB
efficientQ4_K_M18 GB17.04 GB
compressedQ3_K_M14 GB12.36 GB
Scaling

Context Length Impact

KV cache VRAM at Q6_K quality. Longer context = more memory.

ContextKV CacheTotal VRAM
2K205 MB24.2 GBexceeds 24 GB
4K410 MB24.4 GBexceeds 24 GB
8K819 MB24.8 GBexceeds 24 GB
16K1.5 GB25.5 GBexceeds 24 GB
32K3.1 GB27.1 GBexceeds 24 GB
64K6.1 GB30.1 GBexceeds 24 GB
128K12.3 GB36.3 GBexceeds 24 GB

Compatible GPUs

37 devices
NVIDIA Grace Blackwell Ultra GB300Q8_0500 tok/sExcellent
Apple M4 Max (128GB Unified)Q8_084 tok/sExcellent
Apple M4 Max (36GB Unified)Q8_084 tok/sExcellent
Apple M4 Max (64GB Unified)Q8_084 tok/sExcellent
Apple M4 Pro (24GB Unified)Q4_K_M71 tok/sExcellent
Apple M4 Ultra (192GB)Q8_0127 tok/sExcellent
AMD Radeon Pro W7900Q8_0134 tok/sExcellent
NVIDIA GeForce RTX 3090Q5_K_M213 tok/sExcellent
NVIDIA GeForce RTX 4060 Ti 16GBQ3_K_M98 tok/sExcellent
NVIDIA GeForce RTX 4070 Ti SuperQ3_K_M229 tok/sExcellent
NVIDIA GeForce RTX 4080 SuperQ3_K_M251 tok/sExcellent
NVIDIA GeForce RTX 4090Q5_K_M229 tok/sExcellent
NVIDIA RTX 4090 Laptop (150-175W)Q3_K_M175 tok/sExcellent
NVIDIA GeForce RTX 5080Q3_K_M328 tok/sExcellent
NVIDIA GeForce RTX 5090Q8_0278 tok/sExcellent
NVIDIA RTX PRO 6000 BlackwellQ8_0279 tok/sExcellent
NVIDIA RTX PRO 6000 Blackwell Max-QQ8_0279 tok/sExcellent
AMD Radeon RX 7900 XTXQ5_K_M218 tok/sExcellent
AMD Radeon RX 9070Q3_K_M218 tok/sExcellent
AMD Radeon RX 9060 XT 16GBQ3_K_M109 tok/sExcellent
NVIDIA GeForce RTX 5060 Ti 16GBQ3_K_M110 tok/sExcellent
Apple M3 Pro (18GB Unified)Q3_K_M51 tok/sGood
Apple M4 Pro (48GB)Q8_042 tok/sGood
Apple M4 (16GB Unified)Q3_K_M8 tok/sNot viable
NVIDIA GeForce RTX 3060 12GBQ3_K_M8 tok/sNot viable
NVIDIA GeForce RTX 4070 SuperQ3_K_M8 tok/sNot viable
NVIDIA GeForce RTX 4070 Ti 12GBQ3_K_M8 tok/sNot viable
NVIDIA RTX 4080 Laptop (120-150W)Q3_K_M8 tok/sNot viable
Apple M1 (8GB Unified)Q3_K_MNot viable
Apple M1 (16GB Unified)Q3_K_MNot viable
Apple M1 Pro (16GB Unified)Q3_K_MNot viable
Apple M2 (8GB Unified)Q3_K_MNot viable
Apple M2 (16GB Unified)Q3_K_MNot viable
Apple M2 Pro (16GB Unified)Q3_K_MNot viable
Apple M3 (8GB Unified)Q3_K_MNot viable
Apple M3 (16GB Unified)Q3_K_MNot viable
AMD Radeon RX 9060 XT 8GBQ3_K_MNot viable

Showing 37 of 37 entries

Builder Context

Gemma 4 26B-A4B is commonly used with Claude Code, Codex CLI, Continue, Cursor, LM Studio, Ollama, Open WebUI, Windsurf. For an AI coding workflow, pair it with an embedding model like nomic-embed-text for local RAG.

FAQ

Frequently Asked Questions

How much VRAM does Gemma 4 26B-A4B need?
Gemma 4 26B-A4B requires 24 GB VRAM at recommended quality (Q6_K). At lower quality settings, it can fit in as little as 14 GB.
What is the best GPU for Gemma 4 26B-A4B?
The NVIDIA Grace Blackwell Ultra GB300 delivers the best performance for Gemma 4 26B-A4B, achieving 500 tok/s at Q8_0 with an excellent rating.
Can I run Gemma 4 26B-A4B on an RTX 4060 Ti?
Yes. On the NVIDIA GeForce RTX 4060 Ti 16GB, Gemma 4 26B-A4B runs at 98 tok/s (Q3_K_M, excellent).
What quantization should I use for Gemma 4 26B-A4B?
For the best quality, use Q6_K (24 GB VRAM). If your GPU has limited VRAM, Q3_K_M (14 GB) is the most efficient option with acceptable quality.
Is Gemma 4 26B-A4B good for coding?
Yes. Gemma 4 26B-A4B is used with Claude Code, Codex CLI, Continue, Cursor, LM Studio, Ollama, Open WebUI, Windsurf for local AI coding. For the best coding experience, pair it with an embedding model for local RAG.

Related Guides

All models

Data confidence: estimated. Source

VRAM requirements are calculated from model parameters and may vary by inference engine, context length, and batch size. Performance estimates are based on community benchmarks and should be verified for your specific configuration.Gemma is a trademark of its respective owner. OwnRig is not affiliated with or endorsed by the model creator.