Google: Gemma 4 26B A4B (free)
google/gemma-4-26b-a4b-it:free
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
Gemma 4 26B A4B IT is Google's instruction-tuned Mixture-of-Experts model, listed on OpenRouter as a free endpoint. The model card describes 25.2B total parameters with about 3.8B active per token, text and image input for this variant (the catalog also lists video), text output, a 256K-token context window, native function calling, and an optional thinking mode that the catalog reports as disabled by default. Vendor claims of near-4B-class inference speed are source-reported rather than independently verified, and catalog index data suggests limited measured agentic performance.[1][2]
Strengths
- Mixture-of-Experts design activates only a small fraction of its 25.2B parameters per token, which the vendor says enables inference speeds close to a 4B model.[1][2]
- Long 256K-token context window combined with image understanding features such as document parsing, OCR, and variable-resolution image processing.[1][2]
- Native tool and function calling plus a configurable thinking mode, exposed through the endpoint's supported parameters and chat template controls.[1][2]
Limitations
- Catalog-reported index data indicates weak measured results on agentic tasks, and the vendor's own table shows this variant trailing the dense 31B model on multi-needle long-context retrieval.[1][2]
- No audio input is documented for this variant, and the sources disagree on video: the catalog lists it as an input while the model card's variant table lists only text and image.[1][2]
- Evidence is limited to a catalog entry plus vendor-authored documentation; thinking is disabled by default on the endpoint, the knowledge cutoff is unspecified, and real-world behavior should be validated on your own workloads.[1][2]
Where to use it
- Prototyping long-context document and image question answering or OCR pipelines at no listed cost on the free endpoint.[1][2]
- Coding assistance and tool-using assistant workflows where an optional thinking mode and function calling are useful.[1][2]
- Multilingual chat applications, drawing on reported support across more than 35 languages and pretraining on over 140.[2]
Sources
- [1] OpenRouter catalog · Google: Gemma 4 26B A4B (free)Sources checked: Nov 13, 2026
- [2] Model documentation · Google: Gemma 4 26B A4B (free)Sources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...
Specifications
- Model ID
- google/gemma-4-26b-a4b-it:free
- Canonical ID
- google/gemma-4-26b-a4b-it-20260403
- Creator
- Added to OpenRouter
- 03.04.2026
- Context length
- 262,144
- Modality
- text+image+video->text
- Input modalities
- image, text, video
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $0 / out $0
- Knowledge cutoff
- —
- Expiration date
- —
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | No measurement | reasoning: optional · parameters: include_reasoning, reasoning |
| Coding | CC — below the B band | — |
| Agents | CC — below the B band | parameters: tool_choice, tools |
| Writing | No measurement | output: text |
| Reasoning | No measurement | reasoning: optional · parameters: include_reasoning, reasoning |
| Document analysis | No measurement | — |
| Vision | No measurement | input: image |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | input: video |
| Embeddings | No measurement | — |
Evaluations
| Source | Benchmark | Value | Grade | Data retrieved |
|---|---|---|---|---|
| Artificial Analysis | Coding index | 39.3 | CC — below the B band | 13.11.2026 |
| Artificial Analysis | Agentic index | 2.9 | CC — below the B band | 13.11.2026 |
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
| Provider | Context length | Price in /1M | Price out /1M | Latency p50 (30m, s) | Throughput p50 (30m, tok/s) | Uptime (30m) |
|---|---|---|---|---|---|---|
| Google AI Studio | 262,144 | $0 | $0 | — | — | 100% |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.