Tomorrow
← Chips & Compute

SambaNova SN40L

SambaNova · #7 most active of 18 in Chips & Compute

Compare →

Reconfigurable dataflow units with three-tier memory for serving many models per node.

83

Trust score · Mostly verified

3 data points checked · 2026-09-13

Model stack: DisputedEntry price: VerifiedVersion: VerifiedHow this works →

Current version

SN40L

Entry cost

Free tier available. Paid plans include Developer (Pay-as-you-go) and Enterprise (Subscription-based).

Changes / 30d

5

The short answers · verified September 15, 2026

What is the latest version of SambaNova SN40L?
The current shipped version of SambaNova SN40L is SN40L, as of September 15, 2026.
How much does SambaNova SN40L cost?
SambaNova SN40L has no published flat entry price. Free tier available. Paid plans include Developer (Pay-as-you-go) and Enterprise (Subscription-based).
What AI model does SambaNova SN40L use?
SambaNova SN40L runs primarily on DeepSeek-V3.1.

Source: SambaNova — cloud.sambanova.ai · Reusable under CC BY 4.0 — cite Tomorrow

Capabilities

  • three-tier memory
  • many models per node
  • fast open-model serving
  • Reconfigurable Dataflow Unit (RDU)
  • Three-tier memory architecture
  • Dataflow processing
  • Low power inference
  • Multi-model concurrency
  • OpenAI-compatible APIs
  • OpenAI Compatible APIs
  • Fast Inference
  • OpenAI Compatibility
  • Anthropic Compatibility
  • Function calling
  • JSON mode
  • Vision
  • Video
  • Audio
  • Embeddings
  • Prompt caching
  • Text generation
  • Inference
  • OpenAI compatibility
  • RDU
  • Fast inference
  • Three-tier memory
  • JSON Mode
  • Function Calling
  • API Access
  • Dataflow Architecture
  • Vision support
  • Inference Platform
  • OpenAI compatible API
  • API keys
  • OpenAI-compatible API
  • OpenAI-Compatible APIs
  • SambaOrchestrator
  • RDU Acceleration
  • Text Generation
  • Reasoning
  • API
  • API access
  • Custom Checkpoints
  • RAG
  • Anthropic compatibility
  • APIs
  • AI Inference
  • On-Demand Pricing
  • Inference API
  • Text to speech

Pricing

  • Free

    Start exploring SambaNova Cloud. Access to production models with standard rate limits.

    Free

  • Developer (Pay-as-you-go)

    Pay as you go for fast inference, includes Standard Rate Limits.

    Free

Entry price over time

8 of 40 points reconstructed from Internet Archive captures (hollow dots) — dated by capture, not by when we first saw the page. View a capture

9/10/2025 · $09/15/2026 · $0

Geek mode

Models under the hood

  • DeepSeek-V3.1

    DeepSeek

    disclosed

  • gpt-oss-120b

    OpenAI

    disclosed

  • Llama / DeepSeek open weights

    open weights

    disclosed

  • Llama 3.1

    Meta

    disclosed

  • Llama 3.1 (8B, 70B, 405B)

    Meta

    disclosed

  • Llama 3.1 405B

    Meta

    disclosed

  • Llama 3.1 70B

    Meta

    disclosed

  • Llama 3.1 8B

    Meta

    disclosed

  • MiniMax M2.7

    MiniMax

    disclosed

  • RDU dataflow

    SambaNova

    disclosed

  • SambaRack SN40-16

    SambaNova

    inferred

  • SN40L RDU

    SambaNova

    inferred

Context window

—

Public API

yes

Multi-model routing

yes

Stack signals

  • SambaRack
  • SambaStack
  • RDU

Shared model stack

Other tracked products running on the same foundation models — a quick read on how much of the catalog moves when one of these models changes.

green disclosed · amber inferred · grey unattributed. Aliases are folded into one model; provider concentration counts only models with an established vendor.

Reported scale

Reference data. Each figure is whatever the source actually said — weekly users, downloads, revenue run-rate — with its own definition and date. These are not comparable between tools and are never used to rank anything.

No public usage figure on record for this product.

Also in Chips & Compute

All Chips & Compute
Tenstorrent Blackhole

Tenstorrent

16

Open RISC-V AI hardware

Version
Blackhole
Cost
from $999/unit
Model
Tensix cores
Google TPU Ironwood

Google

12

Inference-first TPU pods

Version
Ironwood (7th generation)
Cost
from $5.4/unit
Model
Gemini
Cerebras WSE-3

Cerebras

9

Wafer-scale inference and training

Version
WSE-3
Cost
Free tier / developer tier pricing available (contact/partner integrations)
Model
Codex-Spark
NVIDIA Blackwell

NVIDIA

9

GB200/B200 rack-scale AI systems

Version
Blackwell
Cost
Pricing is not publicly listed (enterprise hardware architecture/system)
Model
Blackwell
AWS Trainium2

AWS

5

Cloud-native training silicon

Version
Trainium2
Cost
Pricing not published; available via AWS EC2 instance pricing.
Model
Neuron compiler
Intel Gaudi 3

Intel

5

Cost-focused training and inference accelerator.

Version
1.24.1
Cost
Pricing not published. Available via Intel Tiber AI Cloud or OEM platforms.
Model
Accelerator

Elsewhere on the same models

Products in other segments that name one of SambaNova SN40L's models in their stack.

ERNIE Bot

BaiduChina

25

Baidu's flagship assistant, now on open-weight ERNIE models.

Version
5.0-thinking-preview
Cost
Pay-as-you-go API pricing based on tokens. Token Plan Personal Edition and some free trials available.
Model
DeepSeek-V3.1

Change history

  • model

    SambaNova SN40L added to its model stack

    DeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, Llama 3.1 (8B, 70B, 405B), Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, MiniMax M2.7, RDU dataflow, SN40L RDU, SambaRack SN40-16, gpt-oss-120bDeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, Llama 3.1 (8B, 70B, 405B), Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, MiniMax M2.7, RDU dataflow, SN40L RDU, SambaRack SN40-16, gpt-oss-120b

    source
  • model

    SambaNova SN40L added to its model stack

    DeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, Llama 3.1 (8B, 70B, 405B), Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, MiniMax M2.7, RDU dataflow, SN40L RDU, SambaRack SN40-16, gpt-oss-120bDeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, Llama 3.1 (8B, 70B, 405B), Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, MiniMax M2.7, RDU dataflow, SN40L RDU, SambaRack SN40-16, gpt-oss-120b

    source
  • capability

    New capabilities: Text to speech

    49 tracked50 tracked · +Text to speech

    source
  • capability

    New capabilities: Inference API

    48 tracked49 tracked · +Inference API

    source
  • capability

    New capabilities: AI Inference, On-Demand Pricing

    46 tracked48 tracked · +AI Inference, On-Demand Pricing

    source
  • model

    SambaNova SN40L added to its model stack

    DeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, Llama 3.1 (8B, 70B, 405B), Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, MiniMax M2.7, RDU dataflow, SN40L RDU, SambaRack SN40-16, gpt-oss-120bDeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, Llama 3.1 (8B, 70B, 405B), Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, MiniMax M2.7, RDU dataflow, SN40L RDU, SambaRack SN40-16, gpt-oss-120b

    source
  • capability

    New capabilities: APIs

    45 tracked46 tracked · +APIs

    source
  • model

    SambaNova SN40L added SambaRack SN40-16 to its model stack

    DeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, Llama 3.1 (8B, 70B, 405B), Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, MiniMax M2.7, RDU dataflow, SN40L RDU, gpt-oss-120bDeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, Llama 3.1 (8B, 70B, 405B), Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, MiniMax M2.7, RDU dataflow, SN40L RDU, SambaRack SN40-16, gpt-oss-120b

    source
  • capability

    New capabilities: Anthropic compatibility

    44 tracked45 tracked · +Anthropic compatibility

    source
  • model

    SambaNova SN40L added to its model stack

    DeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, Llama 3.1 (8B, 70B, 405B), Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, MiniMax M2.7, RDU dataflow, SN40L RDU, gpt-oss-120bDeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, Llama 3.1 (8B, 70B, 405B), Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, MiniMax M2.7, RDU dataflow, SN40L RDU, gpt-oss-120b

    source
  • capability

    New capabilities: Custom Checkpoints, RAG

    42 tracked44 tracked · +Custom Checkpoints, RAG

    source
  • capability

    New capabilities: API access

    41 tracked42 tracked · +API access

    source
  • model

    SambaNova SN40L added SN40L RDU to its model stack

    DeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, Llama 3.1 (8B, 70B, 405B), Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, MiniMax M2.7, RDU dataflow, gpt-oss-120bDeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, Llama 3.1 (8B, 70B, 405B), Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, MiniMax M2.7, RDU dataflow, SN40L RDU, gpt-oss-120b

    source
  • capability

    New capabilities: API

    40 tracked41 tracked · +API

    source
  • capability

    New capabilities: Reasoning

    39 tracked40 tracked · +Reasoning

    source
  • capability

    New capabilities: RDU Acceleration, Text Generation

    37 tracked39 tracked · +RDU Acceleration, Text Generation

    source
  • capability

    New capabilities: SambaOrchestrator

    36 tracked37 tracked · +SambaOrchestrator

    source
  • capability

    New capabilities: OpenAI-Compatible APIs

    35 tracked36 tracked · +OpenAI-Compatible APIs

    source
  • capability

    New capabilities: OpenAI-compatible API

    34 tracked35 tracked · +OpenAI-compatible API

    source
  • version

    SambaNova SN40L moved to SN40L

    SambaRack SN40-16SN40L

    source
  • capability

    New capabilities: Inference Platform, OpenAI compatible API, API keys

    31 tracked34 tracked · +Inference Platform, OpenAI compatible API, API keys

    source
  • version

    SambaNova SN40L moved to SambaRack SN40-16

    SN40LSambaRack SN40-16

    source
  • model

    SambaNova SN40L added Llama 3.1 (8B, 70B, 405B) to its model stack

    DeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, MiniMax M2.7, RDU dataflow, gpt-oss-120bDeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, Llama 3.1 (8B, 70B, 405B), Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, MiniMax M2.7, RDU dataflow, gpt-oss-120b

    source
  • capability

    New capabilities: Vision support

    30 tracked31 tracked · +Vision support

    source
  • capability

    New capabilities: Dataflow Architecture

    29 tracked30 tracked · +Dataflow Architecture

    source
  • capability

    New capabilities: JSON Mode, Function Calling, API Access

    26 tracked29 tracked · +JSON Mode, Function Calling, API Access

    source
  • version

    SambaNova SN40L moved to SN40L

    v2.0.2SN40L

    source
  • capability

    New capabilities: Fast inference, Three-tier memory

    24 tracked26 tracked · +Fast inference, Three-tier memory

    source
  • version

    SambaNova SN40L moved to v2.0.2

    SN40Lv2.0.2

    source
  • capability

    New capabilities: RDU

    23 tracked24 tracked · +RDU

    source
  • capability

    New capabilities: Inference, OpenAI compatibility

    21 tracked23 tracked · +Inference, OpenAI compatibility

    source
  • capability

    New capabilities: Text generation

    20 tracked21 tracked · +Text generation

    source
  • pricing

    Entry price decreased to $0/mo

    $0.1$0

    source
  • model

    SambaNova SN40L added Llama 3.1 8B, Llama 3.1 70B, Llama 3.1 405B to its model stack

    DeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, MiniMax M2.7, RDU dataflow, gpt-oss-120bDeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, MiniMax M2.7, RDU dataflow, gpt-oss-120b

    source
  • capability

    New capabilities: Fast Inference, OpenAI Compatibility, Anthropic Compatibility

    1010

    source
  • version

    SambaNova SN40L moved to SN40L

    SN40-16SN40L

    source
  • version

    SambaNova SN40L moved to SN40-16

    SN40LSN40-16

    source
  • capability

    New capabilities: OpenAI Compatible APIs

    95

    source
  • model

    SambaNova SN40L added DeepSeek-V3.1, Llama 3.1, MiniMax M2.7, gpt-oss-120b to its model stack

    Llama / DeepSeek open weights, RDU dataflowDeepSeek-V3.1, Llama / DeepSeek open weights, Llama 3.1, MiniMax M2.7, RDU dataflow, gpt-oss-120b

    source
  • capability

    New capabilities: Multi-model concurrency, OpenAI-compatible APIs

    76

    source
Subscribe to SambaNova SN40L changes

SambaNova SN40L compared

Straight head-to-head pages against the busiest products in Chips & Compute.

SambaNova SN40L vs Tenstorrent BlackholeSambaNova SN40L vs Google TPU IronwoodSambaNova SN40L vs Cerebras WSE-3SambaNova SN40L vs NVIDIA BlackwellSambaNova SN40L vs AWS Trainium2SambaNova SN40L vs Intel Gaudi 3

How to cite this page

Free to cite and reuse under CC BY 4.0. Permalink: https://tomorrow.aliensquad.ai/tools/sambanova

APA
Tomorrow. (2026). SambaNova SN40L — version, pricing and model stack [Data set entry]. AlienSquad. Retrieved 2026-09-17, from https://tomorrow.aliensquad.ai/tools/sambanova
BibTeX
@misc{tomorrow-tools-sambanova,
  author       = {{Tomorrow}},
  title        = {SambaNova SN40L — version, pricing and model stack},
  year         = {2026},
  publisher    = {AlienSquad},
  howpublished = {\url{https://tomorrow.aliensquad.ai/tools/sambanova}},
  note         = {Accessed: 2026-09-17}
}