MODEL PROOF
Gemma 4 26B A4B
FRONTIER
Nothing is both better and cheaper under this workload.
Serving precision differs between offers.
- Independent LMArena score
- 1,433.9 ± 7.42 · 6,081 votes
- Context window
- 262.14K
- Maximum output
- 235.93K
- Input modalities
- image, text, video
- Output modalities
- text
- Published input price
- $0.069 / 1M tokens
- Published output price
- $0.300 / 1M tokens
- Pricing kind
- fixed
Inspect complete billing conditions and endpoint terms below.
Price from Io Net: the cheapest offer that is serving, at standard delivery and at a declared precision that is not 4-bit (bf16), at its standard rate. The maker’s own precision is not known.
Cheaper right now: $0.059 in · $0.255 out per million tokens at Io Net, a promotion of 15%. The price on this page is Io Net’s standard rate: the ranking uses the standard rate, never a promotion.
Every price dimension · USD per million tokens
| Input | $0.069/M |
|---|---|
| Output | $0.300/M |
| Cached input | $0.029/M |
| Cache write | Unknown/M |
| Cache write 1h | Unknown/M |
| Reasoning | Unknown/M |
Who sells it
The same weights, different shops. Cheapest is not like-for-like when serving precision differs.
Headline cheapest is a lower precision. Like-for-like at the best declared precision is $0.060/M from DekaLLM.
13 provider offers · rates, limits and conditions
Rates are USD per million tokens. Endpoint terms can differ even when the quoted price and precision match. A listed parameter is a provider declaration, not a task-success test.
| Seller | Input | Output | Precision | Uptime (1d) |
|---|---|---|---|---|
| Darkbloomdarkbloom Endpoint terms
| $0.042 | $0.220 | undeclared | 99.99% |
| Io Netio-net/bf16 Endpoint terms
| $0.059 | $0.255 | bf16 | 99.75% |
| DekaLLMdekallm/bf16 Endpoint terms
| $0.060 | $0.330 | bf16 | 99.81% |
| DeepInfradeepinfra/fp8 Endpoint terms
| $0.070 | $0.340 | fp8 | 99.77% |
| Makoramakora Endpoint terms
| $0.080 | $0.320 | undeclared | 98.52% |
| NextBitnextbit/bf16 Endpoint terms
| $0.090 | $0.300 | bf16 | 99.93% |
| CoreWeavecoreweave/bf16 Endpoint terms
| $0.100 | $0.300 | bf16 | 99.97% |
| Cloudflarecloudflare Endpoint terms
| $0.100 | $0.300 | undeclared | 99.86% |
| Venicevenice/bf16 Endpoint terms
| $0.130 | $0.400 | bf16 | 98.84% |
| Parasailparasail/bf16 Endpoint terms
| $0.130 | $0.400 | bf16 | 99.15% |
| Novitanovita/bf16 Endpoint terms
| $0.130 | $0.400 | bf16 | 99.81% |
| SiliconFlowsiliconflow/fp8 Endpoint terms
| $0.140 | $0.400 | fp8 | 95.41% |
| Googlegoogle-vertex/global Endpoint terms
| $0.150 | $0.600 | undeclared | 97.49% |
Independent scores
| LMArena Elo | 1433.9default |
|---|
LMArena Elo under CC BY 4.0. Absence of another board is not a score of zero.
Capability
| Context window | 262K |
|---|---|
| Max output | 235K |
| Input modes | image, text, video |
| Tool use | yes |
| Reasoning | optional |
| Open weights | Unknown |
Provenance
| Price source | openrouter.ai |
|---|---|
| Fetched | 2026-10-06 |
| Quality data | verified |
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...
Questions this page answers
What does Gemma 4 26B A4B cost?
$0.069/M in, $0.300/M out in this catalogue, as of the fetch date on this page. That is the API row, not a subscription.
Does Gemma 4 26B A4B have an independent quality score?
Gemma 4 26B A4B has an LMArena score in this catalogue.
What beats Gemma 4 26B A4B ?
Nothing is both better and cheaper than Gemma 4 26B A4B .
Does this page use Artificial Analysis scores for Gemma 4 26B A4B ?
No. Artificial Analysis figures are not published here. Quality on this page is LMArena Elo where a score exists; otherwise the row is unrated.
Is the cheapest Gemma 4 26B A4B endpoint the same product?
Headline cheapest is a lower precision. Like-for-like at the best declared precision is $0.060/M from DekaLLM.
MODEL MONUMENT
Save or share this proof
Badge
The verdict as one SVG, for a README or a docs page. It states this model's dominance status at the balanced workload on the LMArena lens, with the date it was computed, and it is rebuilt with the catalogue — so it changes when the verdict changes, including to one you would rather it did not.
[](https://undominated.ai/models/google__gemma-4-26b-a4b-it/)<a href="https://undominated.ai/models/google__gemma-4-26b-a4b-it/"><img src="https://undominated.ai/badge/google__gemma-4-26b-a4b-it.svg" alt="Gemma 4 26B A4B — Undominated.ai dominance verdict" height="36"></a>Direct file: https://undominated.ai/badge/google__gemma-4-26b-a4b-it.svg — a static SVG, written by scripts/build-badges.mjs on every build, so the copy you embed is never
older than the last deploy.
