DiffusionGemma suits teams building speed-sensitive local or on-device apps, especially document parsing and OCR, where its bidirectional decoder attention beats standard causal-attention models. The tradeoff: it trails its own autoregressive backbone, Gemma 4, by 5 to 19 points on reasoning benchmarks, so accuracy-critical coding or math work should stick with that sibling instead.
DiffusionGemma is Google DeepMind's open-source discrete text diffusion language model, released June 10, 2026. Instead of predicting one token at a time, it denoises 256-token blocks of text in parallel, running at over 1,000 tokens per second on modern GPUs, with weights free to self-host commercially.
Where it sits
- --$ per 1M tokensBlended price (3:1)Lower is better-- / 61peer median $1.71/Mvendor price, checked by HokAI
- 1008 tok/stokens/sOutput speedHigher is better#2 / 36peer median 90 tok/scited: Artificial Analysis
- --% solvedSWE-bench VerifiedHigher is better-- / 28peer median 78.3%per source, see benchmark scores
- 73.2%% correctGPQA DiamondHigher is better#34 / 44peer median 88.3%per source, see benchmark scores
In the bottom third on GPQA Diamond (rank 34 of 44), and one of 64 whose vendor states it does not train on customer data.
Ranks are against GA models on HokAI that publish the same figure; ties share a rank.
Provider: Google · Family: DiffusionGemma
Context window: 256,000 tokens
Input modalities: text, image, video, tool-calls · Output: text, tool-calls
About DiffusionGemma 26B-A4B
DiffusionGemma 26B-A4B is Google DeepMind's first discrete text diffusion language model, released publicly on June 10, 2026 as an open-weight model. It runs on the same Mixture-of-Experts backbone as Gemma 4 26B-A4B (128 fine-grained experts, top-8 routing, 25.2 billion total parameters with 3.8 billion active per token), but replaces one-token-at-a-time decoding with a parallel denoising process that generates a full block of tokens per forward pass over up to 48 refinement steps. The design targets speed-critical, low-concurrency workloads: inline editing, rapid prototyping, and interactive assistants where raw throughput matters more than chasing the top benchmark score.
On quality benchmarks, DiffusionGemma scores 77.6% on MMLU Pro, 73.2% on GPQA Diamond, 69.1% on AIME 2026 (no tools), and 69.1% on LiveCodeBench v6, with a Codeforces ELO of 1,429. Its autoregressive sibling Gemma 4 26B-A4B beats it on every one of those: 82.6% MMLU Pro, 82.3% GPQA Diamond, 88.3% AIME 2026, a gap of 5 to 19 points depending on the test. The one benchmark DiffusionGemma wins is OmniDocBench 1.5 document parsing, where bidirectional decoder attention gives it a structural edge over causal models at OCR and layout-aware extraction. HumanEval pass@1 with no tools lands at just 11.0%, a known weak spot tied to the block-generation approach struggling with strict left-to-right output formatting.
DiffusionGemma reaches 1,008 tokens per second running FP8 precision, and 1,288 tokens per second on an H200, roughly 4x an equivalently-sized autoregressive model on the same class of hardware at low concurrency. vLLM added native support on day one, the first discrete diffusion LLM to get that treatment, with same-day support in Hugging Face Transformers, MLX, and SGLang. That speed edge is specific to low-concurrency use: in high-QPS cloud serving, autoregressive models can batch requests to saturate compute, which narrows or erases the throughput gap.
Pricing
Open weights, so there's no license fee or per-token charge from Google, only the compute you run it on. Self-hosted cost runs about $0.50/hr on a consumer GPU (18 GB VRAM, NVFP4, 700+ tok/s) up to $2.50/hr on an H100 (28 GB VRAM, FP8, 1,008 tok/s). NVIDIA NIM offers free hosted inference in preview. On an H100, that works out to roughly $0.70 per million tokens processed.
Key Features
- Parallel Block Generation: Generates an entire block of tokens through iterative denoising instead of one token at a time, the core mechanism behind its throughput advantage over standard autoregressive decoding.
- 256,000-Token Context Window: Supports a large context window using a bidirectional encoder-decoder design, enabling full-document analysis and multi-document reasoning in a single request.
- Native Multimodal Input: Accepts text, images, and video in a single prompt, with confirmed support for OCR, PDF parsing, chart comprehension, screen parsing, and video analysis.
- Open Weights on Hugging Face: Weights are available on Hugging Face, Kaggle, and Google Cloud Vertex AI Model Garden under an Apache 2.0 license, with no royalties or attribution required for commercial use.
- Runs on a Single Consumer GPU: The NVFP4-quantized build fits within 18 GB of VRAM, putting it within reach of a single high-end consumer GPU rather than requiring a datacenter card.
- Configurable Thinking Mode: Offers a reasoning mode that can be toggled per request, adding step-by-step chain-of-thought before the final answer on math, code, and complex multi-step tasks.
Pros
- The fastest open-source inference speed available in its parameter class: roughly 4x an equivalently-sized autoregressive model at low concurrency.
- Fully open-source: self-host, fine-tune, or deploy commercially with no royalties or attribution requirements.
- Wins on document-parsing and OCR benchmarks thanks to the bidirectional attention baked into its decoder, unlike standard causal-attention models.
- Small enough to run on a single high-end consumer GPU thanks to NVFP4 quantization, unusual for a 25B+ parameter model.
Cons
- Trails its autoregressive sibling by double-digit points on GPQA Diamond, AIME, and MMLU Pro reasoning benchmarks.
- HumanEval pass@1 without tools is a known weak point, reflecting the block-generation approach's trouble with strict left-to-right output formats.
- No audio input or output; full-precision BF16 needs 50+ GB of VRAM across multiple GPUs.
- Speed advantage shrinks at high concurrency: autoregressive models close the gap through request batching in production serving.
Benchmarks
- MMLU-Pro: 77.6% vendor-reported · 10 Jun 2026 — A harder version of the 57-subject knowledge exam, % correct.
- AIME 2025: 69.1% vendor-reported · 10 Jun 2026 — Competition-level maths problems from the 2025 exam, % solved.
- HumanEval: 11% vendor-reported · 10 Jun 2026 — Small programs that must pass hidden tests, % passing.
- LiveBench: 69.1% vendor-reported · 10 Jun 2026 — A rolling set of fresh questions that cannot have been in training data, % correct.
- GPQA Diamond: 73.2% vendor-reported · 10 Jun 2026 — PhD-level science questions that are hard to search for, % correct.
- Output speed: 1008 tok/s cited: Artificial Analysis — Median tokens written per second as measured by Artificial Analysis.
A benchmark is an exam, not the job. Scores transfer unevenly between tasks, so weigh the one closest to your workload and read every figure with its source.
Frequently Asked Questions
What are DiffusionGemma 26B-A4B's pricing plans in 2026?
DiffusionGemma costs nothing to license: it's fully open-weight, so Google charges no per-token fee and you only pay for the compute you run it on. Self-hosting the NVFP4 build costs roughly $0.50 an hour on a consumer GPU, while the faster FP8 tier costs about $2.50 an hour on a datacenter-class card. NVIDIA also runs a free hosted preview through NVIDIA NIM if you'd rather not manage a GPU yourself.
Can you use DiffusionGemma 26B-A4B without paying?
Yes. It ships fully open-weight, so self-hosting has no usage caps, seat limits, or per-token fees. NVIDIA's hosted NIM endpoint is also free to try, though capacity is limited during its preview.
Which tools compete with DiffusionGemma 26B-A4B in 2026?
DeepSeek V4 offers higher benchmark accuracy if you're running at scale and don't need the throughput advantage. Mistral Medium 3 is a good fit if you'd rather use a low-cost hosted API instead of self-hosting. Both trade DiffusionGemma's inference speed and open licensing for either stronger reasoning scores or zero infrastructure management.
What separates DiffusionGemma 26B-A4B from Gemma 4?
Gemma 4, the autoregressive backbone DiffusionGemma is built from, wins on every general reasoning benchmark: MMLU Pro (82.6% vs 77.6%), GPQA Diamond (82.3%, roughly a nine-point gap), and AIME 2026 (88.3% vs 69.1%). DiffusionGemma's advantage is speed, not accuracy: it generates whole blocks of tokens in parallel instead of one at a time, roughly 4x faster at low concurrency on the same GPU. The one benchmark where DiffusionGemma pulls ahead is document parsing, where its bidirectional attention outperforms Gemma 4 on OCR and layout-heavy extraction.
How do you set up DiffusionGemma 26B-A4B?
Download the weights from Hugging Face and run them with vLLM, which added native support for DiffusionGemma at launch. Pick your precision based on available VRAM: the NVFP4 build is the easiest starting point for a single consumer GPU, while FP8 needs a datacenter-class card. If you'd rather not manage infrastructure, NVIDIA NIM's free hosted preview lets you test the model with an API key and no local setup.
Top Alternatives
- DeepSeek V4: DeepSeek V4 delivers higher benchmark accuracy for teams running at scale. DiffusionGemma trades some of that accuracy for faster, self-hosted open-weight inference instead.
- Mistral Medium 3: Mistral Medium 3 is the simpler choice as a low-cost hosted API with nothing to manage. DiffusionGemma asks for self-hosting in exchange for roughly 4x the throughput.