Gemma‑4‑26B‑A4B‑it is Google’s first open model based on the Mixture‑of‑Experts (MoE) architecture. With a total of 25.2 billion parameters, only a small fraction — between 3.8 and 4 billion — is activated for each token. According to the developers, this efficiency allows the model to achieve approximately 97% of the quality of the dense 31B model at significantly lower computational cost. At release, the model ranks 6th on the Arena AI leaderboard among open models, outperforming competitors that are 20 times larger.
The 26B A4B model is built on 30 layers and uses hybrid attention with a sliding window of 1024 tokens, supporting a context window of 256 thousand tokens. It has multimodal capabilities, handling both text and images exceptionally well. Unlike dense alternatives, the MoE model is specifically optimised for efficient execution of agentic workflows, demonstrating significant progress over Gemma‑3.
For developers, the key advantage of this model is its exceptional deployment efficiency. Community estimates indicate that the model can generate 162 tokens per second on an NVIDIA RTX 4090 accelerator and can run effectively even on memory‑constrained devices. This makes it an ideal choice for complex agentic systems, deep code analysis, and intensive reasoning tasks where a balance between performance and hardware costs is required.
For the developers’ usage recommendations for the model, please refer to this link: https://ai.google.dev/gemma/docs/core/model_card_4?hl=en
| Model Name | Context | Type | GPU | TPS | Tooling | Status | Link |
|---|---|---|---|---|---|---|---|
| google/gemma-4-26B-A4B-it | 263K | Public | 4×4090 tensor |
238.0 | yes | AVAILABLE | chat |
curl https://chat.immers.cloud/v1/endpoints/generate/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer USER_API_KEY" \
--data-binary @- <<"EOF"
{"model": "gemma-4-26b-a4b-it", "messages": [
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Say this is a test"}
], "temperature": 0, "max_tokens": 150
}
EOF
$response = Invoke-WebRequest https://chat.immers.cloud/v1/endpoints/generate/chat/completions `
-Method POST `
-Headers @{
"Authorization" = "Bearer USER_API_KEY"
"Content-Type" = "application/json; charset=utf-8"
} `
-Body ([System.Text.Encoding]::UTF8.GetBytes((@{
model = "gemma-4-26b-a4b-it"
messages = @(
@{ role = "system"; content = "You are a helpful assistant." },
@{ role = "user"; content = "Say this is a test" })
} | ConvertTo-Json -Depth 10)))
([System.Text.Encoding]::UTF8.GetString($response.RawContentStream.ToArray()) | ConvertFrom-Json).choices[0].message.content
#!pip install OpenAI --upgrade
from openai import OpenAI
client = OpenAI(
api_key="USER_API_KEY",
base_url="https://chat.immers.cloud/v1/endpoints/generate/",
)
chat_response = client.chat.completions.create(
model="gemma-4-26b-a4b-it",
messages=[
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Say this is a test"},
]
)
print(chat_response.choices[0].message.content)
# linux
export ANTHROPIC_BASE_URL=https://chat.immers.cloud/v1/endpoints/generate/
export ANTHROPIC_API_KEY=USER_API_KEY
export ANTHROPIC_MODEL=gemma-4-26b-a4b-it
# windows (ps)
[Environment]::SetEnvironmentVariable("ANTHROPIC_BASE_URL", "https://chat.immers.cloud/v1/endpoints/generate/", "User")
[Environment]::SetEnvironmentVariable("ANTHROPIC_AUTH_TOKEN", "USER_API_KEY", "User")
[Environment]::SetEnvironmentVariable("ANTHROPIC_MODEL", "gemma-4-26b-a4b-it", "User")
# linux
export OPENAI_BASE_URL=https://chat.immers.cloud/v1/endpoints/generate/
export OPENAI_API_KEY=USER_API_KEY
export OPENAI_MODEL=gemma-4-26b-a4b-it
# windows (ps)
[Environment]::SetEnvironmentVariable("OPENAI_BASE_URL", "https://chat.immers.cloud/v1/endpoints/generate/", "User")
[Environment]::SetEnvironmentVariable("OPENAI_API_KEY", "USER_API_KEY", "User")
[Environment]::SetEnvironmentVariable("OPENAI_MODEL", "gemma-4-26b-a4b-it", "User")
An endpoint is a server with pre-installed software that automatically downloads neural network weights and launches an OpenAI-compatible API to handle incoming requests.
A private endpoint is created within the user's project, belongs exclusively to the user, and is billed per server configuration in accordance with the infrastructure pricing.
We recommend creating private endpoints when you need to:
| Name | GPU | TPS | Max Concurrency | |||
|---|---|---|---|---|---|---|
262,144.0 tensor |
2 | $0.93 | 3.361 | Launch | ||
262,144.0 pipeline |
3 | $1.06 | 2.352 | Launch | ||
262,144.0 tensor |
4 | $1.26 | 2.805 | Launch | ||
262,144.0 tensor |
2 | $1.56 | 168.50 | 3.411 | Launch | |
262,144.0 tensor |
2 | $1.56 | 84.00 | 3.411 | Launch | |
262,144.0 tensor |
2 | $1.92 | 249.00 | 3.455 | Launch | |
262,144.0 |
1 | $2.37 | 122.66 | 8.627 | Launch | |
262,144.0 |
1 | $3.83 | 6.366 | Launch | ||
262,144.0 |
1 | $4.11 | 7.904 | Launch | ||
262,144.0 tensor |
2 | $4.61 | 166.86 | 19.947 | Launch | |
262,144.0 |
1 | $4.74 | 13.069 | Launch | ||
262,144.0 tensor |
2 | $9.40 | 28.424 | Launch | ||
| Name | GPU | TPS | Max Concurrency | |||
|---|---|---|---|---|---|---|
262,144.0 tensor |
4 | $1.26 | 1.417 | Launch | ||
262,144.0 pipeline |
3 | $1.34 | 3.181 | Launch | ||
262,144.0 tensor |
2 | $1.56 | 141.00 | 1.580 | Launch | |
262,144.0 tensor |
2 | $1.56 | 1.580 | Launch | ||
262,144.0 tensor |
4 | $1.57 | 4.025 | Launch | ||
262,144.0 tensor |
2 | $1.92 | 163.33 | 1.795 | Launch | |
262,144.0 |
1 | $2.37 | 114.60 | 7.123 | Launch | |
262,144.0 tensor |
2 | $2.93 | 3.575 | Launch | ||
262,144.0 |
1 | $3.83 | 132.83 | 4.912 | Launch | |
262,144.0 |
1 | $4.11 | 209.64 | 6.809 | Launch | |
262,144.0 tensor |
2 | $4.61 | 17.912 | Launch | ||
262,144.0 |
1 | $4.74 | 11.300 | Launch | ||
262,144.0 tensor |
2 | $9.40 | 26.625 | Launch | ||
| Name | GPU | TPS | Max Concurrency | |||
|---|---|---|---|---|---|---|
262,144.0 tensor |
4 | $1.62 | 2.341 | Launch | ||
262,144.0 pipeline |
6 | $1.65 | 2.051 | Launch | ||
262,144.0 pipeline |
3 | $2.29 | 1.014 | Launch | ||
262,144.0 |
1 | $2.37 | 100.14 | 3.629 | Launch | |
262,144.0 pipeline |
3 | $2.83 | 1.290 | Launch | ||
262,144.0 tensor |
4 | $2.89 | 199.00 | 2.438 | Launch | |
262,144.0 tensor |
4 | $3.01 | 2.438 | Launch | ||
262,144.0 tensor |
4 | $3.60 | 238.00 | 2.536 | Launch | |
262,144.0 |
1 | $3.83 | 2.606 | Launch | ||
262,144.0 |
1 | $4.11 | 171.47 | 4.314 | Launch | |
262,144.0 pipeline |
3 | $4.34 | 4.009 | Launch | ||
262,144.0 tensor |
2 | $4.61 | 15.161 | Launch | ||
262,144.0 |
1 | $4.74 | 9.379 | Launch | ||
262,144.0 tensor |
4 | $5.74 | 5.263 | Launch | ||
262,144.0 tensor |
2 | $9.40 | 24.592 | Launch | ||
Contact our dedicated neural networks support team at support@immers.cloud or send your request to the sales department at sale@immers.cloud.