BroadRouter

Gemma 4 26B A4B API pricing

Updated

Among the 11 providers we checked (prices read 2026-11-14 to 2026-11-15), 2 tie for the lowest blended standard price for Gemma 4 26B A4B; DeepInfra charges USD 0.07 per 1M input tokens and USD 0.34 per 1M output tokens. The median across 11 standard prices is USD 0.13 / USD 0.40. Each price is read from the provider's own pricing page.

Google · slug gemma-4-26b-a4b-it · up to 262,144 tokens of context · 11 providers with a standard price. Every price is read from the provider's own pricing page and linked in its row, checked 2026-11-15. No aggregator feed. Download all prices (every provider and model, with sources, CSV or JSON). How it compares with the market: the LLM API price index. Archived list prices since 2026-06-01 (6 archived price points, 0 list-price changes): LLM API price history.

Does Google sell Gemma 4 26B A4B directly?

Google publishes no price for Gemma 4 26B A4B on its own API that we could find, so this page compares hosts only.

Gemma 4 26B A4B model facts

Facts from Google's model card for google/gemma-4-26B-A4B-it and the Hugging Face model API, read 2026-11-16. BroadNet did not verify them, and hosts may serve the model at a different precision. BroadRouter is BroadNet's site.

Gemma 4 26B A4B: facts as stated by the model card and the Hugging Face model API, with sources
FactAs statedSource
Open weightsYes. The weights are in a public Hugging Face repository (google/gemma-4-26B-A4B-it) with no access gate.Hugging Face model API, checked 2026-11-16
Quote
“"private": false, "gated": false, 2 .safetensors files listed”
LicenceApache 2.0 (apache-2.0)Google model card metadata, checked 2026-11-16
Quote
“license: apache-2.0”
Total parameters25.2BGoogle model card, checked 2026-11-16
Quote
“Total Parameters | 25.2B”
Active parameters per token3.8BGoogle model card, checked 2026-11-16
Quote
“Active Parameters | 3.8B”
Context length256K tokensGoogle model card, checked 2026-11-16
Quote
“Context Length | 256K tokens”
Input and outputText and image in; text outGoogle model card, checked 2026-11-16
Quote
“Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output.”
Tool callingYesGoogle model card, checked 2026-11-16
Quote
“Function Calling – Native support for structured tool use”
First published on Hugging Face (not necessarily the release date)2026-03-11Hugging Face model API, checked 2026-11-16
Quote
“"createdAt": "2026-03-11T21:25:57.000Z"”

Gemma 4 26B A4B price per provider

USD per 1M tokens. Standard prices are ranked on a blend of 3 input tokens to 1 output token, cheapest first; the dearest standard price is 91% above the cheapest. Free rows, the maker's serving version and other service tiers follow, tagged and unranked.

Gemma 4 26B A4B API prices by provider
#ProviderInputCached inputOutputBlended 3:1ContextPrecisionSource
1DeepInfra tied cheapest
google/gemma-4-26B-A4B-it
Notes and source quote
“$0.07 in $0.34 out / 1M tokens”
USD 0.07not publishedUSD 0.34USD 0.1375262,144FP8DeepInfra page
checked 2026-11-14
re-checked 2026-11-16
2Featherless tied cheapest
google/gemma-4-26B-A4B-it
Notes and source quote

Developer plan (per-token billing, from 50 USD a month in credits).

“PRICING: INPUT $0.07 / CACHED $0.05 / OUTPUT $0.34”
USD 0.07USD 0.05USD 0.34USD 0.137532,768FP8Featherless page
checked 2026-11-15
re-checked 2026-11-16
3Cloudflare Workers AI
@cf/google/gemma-4-26b-a4b-it
Notes and source quote

Cloudflare publishes this USD per-token price alongside the neuron price.

“@cf/google/gemma-4-26b-a4b-it $0.100 per M input tokens $0.300 per M output tokens”
USD 0.10not publishedUSD 0.30USD 0.15not statednot statedCloudflare Workers AI page
checked 2026-11-14
re-checked 2026-11-16
4CoreWeave Forge
google/gemma-4-26B-A4B-it
Notes and source quote
“Google Gemma 4 26B A4B Instruct $0.10 $0.30 $0.05”
USD 0.10USD 0.05USD 0.30USD 0.15262,144not statedCoreWeave Forge page
checked 2026-11-15
re-checked 2026-11-16
5Amazon Bedrock
Notes and source quote

Table headed "Regions: US East (N. Virginia), US East (Ohio) and US West (Oregon)".

“Gemma 4 26B A4B $0.13 $0.40”
USD 0.13not publishedUSD 0.40USD 0.1975not statednot statedAmazon Bedrock page
checked 2026-11-14
re-checked 2026-11-16
6GMI Cloud
google/gemma-4-26b-a4b-it
Notes and source quote

No cache-read price published.

“Price (per 1M tokens) Input $0.13 Output $0.40”
USD 0.13not publishedUSD 0.40USD 0.1975256,000FP8GMI Cloud page
checked 2026-11-15
re-checked 2026-11-16
7Novita AI
google/gemma-4-26b-a4b-it
Notes and source quote
“Gemma 4 26B A4B 256K $0.13 /Mt $0.4 /Mt”
USD 0.13not publishedUSD 0.40USD 0.1975262,144not statedNovita AI page
checked 2026-11-14
re-checked 2026-11-16
8Parasail
Notes and source quote
“Gemma 4 26B-A4B $0.13 $0.40 $0.05”
USD 0.13USD 0.05USD 0.40USD 0.1975not statednot statedParasail page
checked 2026-11-14
re-checked 2026-11-16
9Venice
google-gemma-4-26b-a4b-it
Notes and source quote
“| Google Gemma 4 26B A4B Instruct | `google-gemma-4-26b-a4b-it` | $0.13 | $0.40 | $0.05 | — | Private |”
USD 0.13USD 0.05USD 0.40USD 0.1975256,000BF16Venice page
checked 2026-11-15
re-checked 2026-11-16
10SiliconFlow
Notes and source quote

Context shown as 262K.

“gemma-4-26B-A4B-it 262K $0.14 $0.05 $0.4”
USD 0.14USD 0.05USD 0.40USD 0.205262,144not statedSiliconFlow page
checked 2026-11-14
re-checked 2026-11-16
11Google Vertex AI
gemma-4-26b
Notes and source quote

Vertex AI managed API.

“Gemma 4 26B Input $0.15 … Output $0.60 … Cache Hit $0.015”
USD 0.15USD 0.015USD 0.60USD 0.2625not statednot statedGoogle Vertex AI page
checked 2026-11-14
re-checked 2026-11-16
—DeepInfra flex / spot
google/gemma-4-26B-A4B-it
Notes and source quote

Flex service tier (0.8× standard).

“Flex (0.8×) $0.056 $0.272”
USD 0.056not publishedUSD 0.272USD 0.11262,144FP8DeepInfra page
checked 2026-11-14
re-checked 2026-11-16
—DeepInfra priority
google/gemma-4-26B-A4B-it
Notes and source quote

Priority service tier (1.5× standard).

“Priority (1.5×) $0.105 $0.51”
USD 0.105not publishedUSD 0.51USD 0.2062262,144FP8DeepInfra page
checked 2026-11-14
re-checked 2026-11-16
—BroadNet Inference this site's owner
broadnet/gemma-4-26b-a4b
BroadNet Gemma 4 VL 26B. Not ranked: BroadRouter is BroadNet's own site. Model page
USD 0.10not publishedUSD 0.22USD 0.1332,768not statedpricing.json
synced 2026-11-14

Report a wrong price or number (email to info@broadnet.ai, with this page's link filled in).

What does Gemma 4 26B A4B cost for a real workload?

Monthly cost over 30 days at each standard price, using the presets of the LLM API cost calculator. Providers without a cached price bill cached input at the input price.

  • Support chatbot: 2,000 requests a day, 1,500 input tokens (60% cached), 300 output.
  • RAG over documents: 1,000 requests a day, 8,000 input tokens (30% cached), 500 output.
  • Coding agent (one developer): 1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning.
Monthly cost of three workloads on Gemma 4 26B A4B, by provider
ProviderSupport chatbotRAG over documentsCoding agent (one developer)
DeepInfraUSD 12.42USD 21.90USD 156.60
FeatherlessUSD 11.34USD 20.46exceeds context
Cloudflare Workers AIUSD 14.40USD 28.50USD 207.00
CoreWeave ForgeUSD 11.70USD 24.90USD 130.50
Amazon BedrockUSD 18.90USD 37.20USD 270.00
GMI CloudUSD 18.90USD 37.20USD 270.00
Novita AIUSD 18.90USD 37.20USD 270.00
ParasailUSD 14.58USD 31.44USD 147.60
VeniceUSD 14.58USD 31.44USD 147.60
SiliconFlowUSD 14.94USD 33.12USD 150.30
Google Vertex AIUSD 17.01USD 35.28USD 117.45
BroadNet Inference (this site's owner)USD 12.96USD 27.30exceeds context

For the support chatbot, the cheapest price (Featherless, USD 11.34 a month) saves USD 3.24 against the median (USD 14.58). For the coding agent the gap is USD 32.85 a month (22% lower). The winner changes with the workload: Google Vertex AI wins the cache-heavy agent, Featherless the chatbot.

Which Gemma 4 26B A4B provider is cheapest for your traffic?

DeepInfra and Featherless have the lowest input price (USD 0.07) and Cloudflare Workers AI and CoreWeave Forge the lowest output price (USD 0.30). They cost the same at 1.3 input tokens per output token. With more input than that, as with long prompts or retrieval, DeepInfra and Featherless are cheaper; with less, as with long answers, Cloudflare Workers AI and CoreWeave Forge are.

Gemma 4 26B A4B: price per 1M tokens at 3 input to 1 output, by share of input served from cache
ProviderNo cache hits50% cached90% cached
DeepInfra (no cache price)USD 0.1375 cheapestUSD 0.1375USD 0.1375
FeatherlessUSD 0.1375 cheapestUSD 0.13 cheapestUSD 0.124
Cloudflare Workers AI (no cache price)USD 0.15USD 0.15USD 0.15
CoreWeave ForgeUSD 0.15USD 0.1313USD 0.1163 cheapest
Amazon Bedrock (no cache price)USD 0.1975USD 0.1975USD 0.1975
GMI Cloud (no cache price)USD 0.1975USD 0.1975USD 0.1975

USD per 1M tokens, 3 input tokens to 1 output token, each provider's cheapest standard row; 5 dearer providers are in the price table above. A provider without a cached price bills cached input at the input price. Any charge for writing to the cache, which some providers add, is not included.

6 of 11 providers publish a cached-input price below the input price: Featherless (29% off), CoreWeave Forge (50% off), Parasail (62% off), Venice (62% off), SiliconFlow (64% off), Google Vertex AI (90% off). DeepInfra, Cloudflare Workers AI, Amazon Bedrock, GMI Cloud and Novita AI publish none, so cached tokens there bill at the input price.

Context windows differ: 262,144 tokens at DeepInfra, CoreWeave Forge, Novita AI and SiliconFlow; 256,000 tokens at GMI Cloud and Venice; 32,768 tokens at Featherless. Cloudflare Workers AI, Amazon Bedrock, Parasail and Google Vertex AI do not state one. The cheapest row is not always the one that fits a long prompt.

Gemma 4 26B A4B on BroadNet Inference

BroadRouter is BroadNet's own site. BroadNet Inference serves BroadNet Gemma 4 VL 26B as broadnet/gemma-4-26b-a4b at USD 0.10 input and USD 0.22 output per 1M tokens, with a 32,768-token context. On the 3:1 blend that is at or below every standard price above. It also offers zero data retention, one OpenAI-compatible key for every BroadNet model and a flat USD 0.50 hourly key.

BroadNet Gemma 4 VL 26B on BroadNet: setup, limits and code

What does Gemma 4 26B A4B cost in other languages?

The same text needs a different number of tokens in each language. With the Gemma 4 VL 26B tokenizer, sending 1M characters costs this much at the cheapest standard input price we found for Gemma 4 26B A4B:

ContentInput per 1M charactersCheapest provider
EnglishUSD 0.015DeepInfra
Arabic MSAUSD 0.024DeepInfra
HindiUSD 0.019DeepInfra
ChineseUSD 0.047DeepInfra
CodeUSD 0.020DeepInfra

From BroadRouter's real cost by language dataset, which covers 15 languages and content types for every model we could match to a measured tokenizer.

Published benchmark results

BroadNet did not run these benchmarks; results are as published by the Google model card and Epoch AI and may use different settings.

Published benchmark results for Gemma 4 26B A4B, with sources
BenchmarkResultSetting or noteSource
MMLU Pro82.6%Instruction-tuned modelGoogle model card, checked 2026-11-16
GPQA Diamond82.3%Instruction-tuned modelGoogle model card, checked 2026-11-16
LiveCodeBench v677.1%Instruction-tuned modelGoogle model card, checked 2026-11-16
AIME 2026 no tools88.3%Instruction-tuned modelGoogle model card, checked 2026-11-16
MMMU Pro73.8%Instruction-tuned modelGoogle model card, checked 2026-11-16
MATH-Vision82.4%Instruction-tuned modelGoogle model card, checked 2026-11-16
GPQA Diamond0.732Minimal reasoning setting; Epoch run gemma-4-26b-a4b-it_minimal, mean score on a 0 to 1 scaleEpoch AI, checked 2026-11-16
OTIS Mock AIME 2024-20250.822Minimal reasoning setting; Epoch run gemma-4-26b-a4b-it_minimal, mean score on a 0 to 1 scaleEpoch AI, checked 2026-11-16

Epoch AI rows: Epoch AI, Capabilities & benchmarking, licensed CC BY 4.0, shown as the mean score on a 0 to 1 scale, rounded to three decimals (hover a value for the exact figure in Epoch's data file).

Estimate your own bill

The Google models whose cheapest standard price is closest to Gemma 4 26B A4B's (USD 0.1375 blended):

Google models priced closest to Gemma 4 26B A4B
ModelCheapest inputCheapest outputMax contextProviders
gemma-4-31B-it-turboUSD 0.09USD 0.34262,1442
Gemma 4 31BUSD 0.10USD 0.34262,14412
Gemini 2.5 Flash-LiteUSD 0.10USD 0.40—2
Gemma 3 27BUSD 0.08USD 0.16198,0006
Gemma 3 12BUSD 0.05USD 0.15131,0724
Gemma 3 4BUSD 0.04USD 0.08131,0723

Priced closest from other makers: GPT-5 Nano (OpenAI, USD 0.1375) · MiMo V2.6 Flash (Xiaomi, USD 0.14) · Qwen3-30B-A3B-Instruct-2507 (Qwen, USD 0.1425).

Compare Gemma 4 26B A4B with

Also compared: vs Gemma 4 31B · vs Qwen3.6 35B A3B · vs GPT OSS 20B · vs DeepSeek V4.1 Flash 1017 · vs Llama 3.3 70B · vs GLM 5.3 Flash · vs DeepSeek V4 Flash 0731 · vs OpenAI GPT OSS 120B · vs GPT-5 Nano · vs MiMo V2.6 Flash · vs Qwen3-30B-A3B-Instruct-2507 · vs GLM 4.7 Flash · vs Qwen3 32B FP8.

Every Gemma 4 26B A4B price comparison · All model-vs-model price comparisons

All 429 models: API prices compared · Cheapest LLM APIs right now · LLM API price index (open data) · Download all prices (CSV, JSON) · Price change log

Questions

How much does the Gemma 4 26B A4B API cost per token?

From USD 0.07 per 1M input tokens (USD 0.00000007 per token) and USD 0.30 per 1M output tokens (USD 0.0000003 per token) at the lowest standard prices we found, which can come from different providers. The median of 11 standard prices is USD 0.13 input and USD 0.40 output. Each price was read from the provider's own pricing page; the source is linked in its row.

Which provider is cheapest for Gemma 4 26B A4B?

Among the 11 providers we checked (prices read 2026-11-14 to 2026-11-15), 2 are tied for the lowest price: DeepInfra and Featherless, each USD 0.1375 per 1M on a blend of 3 input tokens to 1 output token. For input alone the lowest price is USD 0.07 (DeepInfra and 1 other, tied); for output alone, USD 0.30 (Cloudflare Workers AI and 1 other, tied). Batch, flex, priority and fast tiers are shown in the table but not ranked.

Does Google sell Gemma 4 26B A4B on its own API?

Google publishes no price for Gemma 4 26B A4B on its own API that we could find, so there is no maker price to compare. The lowest-priced host we checked is DeepInfra at USD 0.07 / USD 0.34 per 1M tokens.

Does Gemma 4 26B A4B have cached input pricing?

6 of 11 standard prices include one. The lowest is USD 0.015 per 1M at Google Vertex AI; the median is USD 0.05. Providers without one bill cached input at the input price. What cached input means.

What is the Gemma 4 26B A4B context window?

Up to 262,144 tokens. 3 standard prices come with less, down to 32,768 tokens (Featherless, GMI Cloud and Venice). Check the context column before you pick the cheapest row for long prompts. Context window calculator.

How much does a coding agent cost per month on Gemma 4 26B A4B?

For 1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning, over 30 days: USD 117.45 a month at Google Vertex AI, USD 150.30 at the median price and USD 270.00 at Novita AI. Try your own numbers.

Is there a batch, flex or priority price for Gemma 4 26B A4B?

Yes: flex / spot from DeepInfra at USD 0.056 / USD 0.272; priority from DeepInfra at USD 0.105 / USD 0.51 per 1M input / output tokens. These tiers trade speed or availability for price (batch runs within hours, flex and spot can be slow or refused at capacity, priority and fast cost more for speed or reliability), so they are shown but not ranked.

Is Gemma 4 26B A4B free to use through an API?

No provider in our data lists a free price for Gemma 4 26B A4B as of 2026-11-15. New BroadNet Inference accounts start with USD 5 of credit and no card, and the credit can be spent on broadnet/gemma-4-26b-a4b at the listed price (rate card synced 2026-11-14). That is a starting credit, not a free price. Free tiers and credits by provider (checked 2026-11-14) lists general offers that do not name a model; all free LLM API models are on one page.

Is Gemma 4 26B A4B open source, with open weights, and what is its licence?

Google publishes the weights on Hugging Face as google/gemma-4-26B-A4B-it, in a public repository with no access gate. The licence, as the Google model card metadata states it: Apache 2.0 (apache-2.0). Open weights means you can download and run the model yourself; what you may do with it, including commercial use, is set by that licence, so read it in full (checked 2026-11-16). These facts come from Google's card and the Hugging Face model API; BroadNet did not verify them, and hosts may serve the model at a different precision.

How many parameters does Gemma 4 26B A4B have?

Total parameters, as the Google model card states them: 25.2B. Active parameters per token: 3.8B. (checked 2026-11-16; BroadNet did not verify these figures.)

Does Gemma 4 26B A4B support tool calling?

Yes, according to the Google model card: “Function Calling – Native support for structured tool use” (checked 2026-11-16). Whether the tools parameter works through an API also depends on the provider serving it, so check that provider's docs. BroadNet did not test it.

Can I use Gemma 4 26B A4B on BroadNet Inference?

Yes. BroadNet Inference serves BroadNet Gemma 4 VL 26B as broadnet/gemma-4-26b-a4b at USD 0.10 input and USD 0.22 output per 1M tokens (rate card synced 2026-11-14), with zero data retention. On the 3:1 blend that is at or below every standard price on this page. BroadRouter is BroadNet's own site. BroadNet Gemma 4 VL 26B on BroadNet.

How do I call Gemma 4 26B A4B with the OpenAI SDK?

On BroadNet Inference, point the OpenAI SDK at https://inference.broadnet.ai/v1 with your BroadNet key and set the model to broadnet/gemma-4-26b-a4b: client = OpenAI(base_url="https://inference.broadnet.ai/v1", api_key=...), then client.chat.completions.create(model="broadnet/gemma-4-26b-a4b", messages=[...]). BroadRouter is BroadNet's own site. Setup notes for agents: llms.txt for Gemma 4 26B A4B on BroadNet.

How does Gemma 4 26B A4B score on benchmarks?

As published, the Google model card reports 82.6% on MMLU Pro and 82.3% on GPQA Diamond; Epoch AI measured 0.732 on GPQA Diamond in its own run, as a mean score on a 0 to 1 scale (checked 2026-11-16). BroadNet did not run these benchmarks, and sources may use different settings, so the figures are not directly comparable. Every published result we found, with its setting and source.

Read as MarkdownAsk ChatGPTAsk ClaudeAsk Perplexity