Tomorrow
← Chips & Compute

Groq LPU

Groq · #8 most active of 18 in Chips & Compute

Compare →

SRAM-based language processing units sold as a token-priced inference cloud.

67

Trust score · Mixed

3 data points checked · 2026-09-13

Model stack: VerifiedEntry price: DisputedVersion: DisputedHow this works →

Current version

LPU

Entry cost

Pricing not published. Contact sales or check account console for usage-based rates.

Changes / 30d

4

The short answers · verified September 15, 2026

What is the latest version of Groq LPU?
The current shipped version of Groq LPU is LPU, as of September 15, 2026.
How much does Groq LPU cost?
Groq LPU has no published flat entry price. Pricing not published. Contact sales or check account console for usage-based rates.
What AI model does Groq LPU use?
Groq LPU runs primarily on canopylabs/orpheus-arabic-saudi.

Source: Groq — console.groq.com · Reusable under CC BY 4.0 — cite Tomorrow

Capabilities

  • deterministic latency
  • OpenAI-compatible API
  • open-model catalogue
  • batch API
  • Fast LLM inference
  • OpenAI Compatibility
  • Prompt Caching
  • Speech to Text
  • Text to Speech
  • OCR and Image Recognition
  • MCP Connectors
  • Tool Use
  • Batch Processing
  • Fast LLM Inference
  • OpenAI-Compatible API
  • Structured Outputs
  • Batch API
  • Reasoning
  • OpenAI API compatibility
  • Model Context Protocol (MCP) Connectors
  • Text to Speech (Orpheus)
  • Web Search
  • Code Execution
  • Text Generation
  • Content Moderation
  • LPU Inference
  • OpenAI-compatible
  • OCR
  • Image Recognition
  • Orpheus OCR
  • Fast Inference
  • Orpheus OCR and Image Recognition

Pricing

  • GroqCloud Console Free Tier / Pay-As-You-Go

    Access to open weights models with rate limits. Developer models have free allocations and pay-as-you-go tiers.

    Free

Entry price over time

1 of 18 points reconstructed from Internet Archive captures (hollow dots) — dated by capture, not by when we first saw the page. View a capture

7/1/2024 · $09/15/2026 · $0

Geek mode

Models under the hood

  • canopylabs/orpheus-arabic-saudi

    Canopy Labs

    disclosed

  • canopylabs/orpheus-v1-english

    Canopy Labs

    disclosed

  • kimi-k2-instruct-0905

    Moonshot AI

    disclosed

  • Llama / Kimi / GPT-OSS

    open weights

    disclosed

  • Llama 3.1 8B Instant

    Meta

    disclosed

  • Llama 3.3 70B Versatile

    Meta

    disclosed

  • LPU v2

    Groq

    disclosed

  • meta-llama/llama-4-maverick-17b-128e-instruct

    Meta

    disclosed

  • meta-llama/llama-4-scout-17b-16e-instruct

    Meta

    disclosed

  • minimax-m2.5

    MiniMax

    disclosed

  • minimax/minimax-m2.5

    MiniMax

    disclosed

  • minimaxai/minimax-m2.5

    MiniMax

    disclosed

  • moonshotai/kimi-k2-instruct-0905

    Moonshot AI

    disclosed

  • openai/gpt-oss-120b

    OpenAI

    disclosed

  • openai/gpt-oss-20b

    OpenAI

    disclosed

  • openai/gpt-oss-safeguard-20b

    OpenAI

    disclosed

  • Orpheus English

    Canopy Labs

    disclosed

  • orpheus-arabic-saudi

    Canopy Labs

    disclosed

  • orpheus-v1-english

    Canopy Labs

    disclosed

  • Qwen 3.6 27B

    Alibaba

    disclosed

  • qwen/qwen3-vl-32b-instruct

    Qwen

    disclosed

  • qwen3-vl-32b-instruct

    Qwen

    disclosed

  • Whisper V3 Large

    OpenAI

    disclosed

  • whisper-large-v3

    OpenAI

    disclosed

Context window

131k

Public API

yes

Multi-model routing

yes

Stack signals

  • python
  • typescript

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

Change history

  • capability

    New capabilities: Orpheus OCR and Image Recognition

    31 tracked32 tracked · +Orpheus OCR and Image Recognition

    source
  • capability

    New capabilities: Fast Inference

    30 tracked31 tracked · +Fast Inference

    source
  • capability

    New capabilities: Orpheus OCR

    29 tracked30 tracked · +Orpheus OCR

    source
  • version

    Groq LPU moved to LPU

    LPU / GroqCloud APILPU

    source
  • model

    Groq LPU added meta-llama/llama-4-scout-17b-16e-instruct, moonshotai/kimi-k2-instruct-0905 to its model stack

    LPU v2, Llama / Kimi / GPT-OSS, Llama 3.1 8B Instant, Llama 3.3 70B Versatile, Orpheus English, Qwen 3.6 27B, Whisper V3 Large, canopylabs/orpheus-arabic-saudi, canopylabs/orpheus-v1-english, kimi-k2-instruct-0905, meta-llama/llama-4-maverick-17b-128e-instruct, minimax-m2.5, minimax/minimax-m2.5, minimaxai/minimax-m2.5, openai/gpt-oss-120b, openai/gpt-oss-20b, openai/gpt-oss-safeguard-20b, orpheus-arabic-saudi, orpheus-v1-english, qwen/qwen3-vl-32b-instruct, qwen3-vl-32b-instruct, whisper-large-v3LPU v2, Llama / Kimi / GPT-OSS, Llama 3.1 8B Instant, Llama 3.3 70B Versatile, Orpheus English, Qwen 3.6 27B, Whisper V3 Large, canopylabs/orpheus-arabic-saudi, canopylabs/orpheus-v1-english, kimi-k2-instruct-0905, meta-llama/llama-4-maverick-17b-128e-instruct, meta-llama/llama-4-scout-17b-16e-instruct, minimax-m2.5, minimax/minimax-m2.5, minimaxai/minimax-m2.5, moonshotai/kimi-k2-instruct-0905, openai/gpt-oss-120b, openai/gpt-oss-20b, openai/gpt-oss-safeguard-20b, orpheus-arabic-saudi, orpheus-v1-english, qwen/qwen3-vl-32b-instruct, qwen3-vl-32b-instruct, whisper-large-v3

    source
  • capability

    New capabilities: OCR, Image Recognition

    27 tracked29 tracked · +OCR, Image Recognition

    source
  • capability

    New capabilities: OpenAI-compatible

    26 tracked27 tracked · +OpenAI-compatible

    source
  • capability

    New capabilities: LPU Inference

    25 tracked26 tracked · +LPU Inference

    source
  • model

    Groq LPU added orpheus-arabic-saudi, qwen3-vl-32b-instruct to its model stack

    LPU v2, Llama / Kimi / GPT-OSS, Llama 3.1 8B Instant, Llama 3.3 70B Versatile, Orpheus English, Qwen 3.6 27B, Whisper V3 Large, canopylabs/orpheus-arabic-saudi, canopylabs/orpheus-v1-english, kimi-k2-instruct-0905, meta-llama/llama-4-maverick-17b-128e-instruct, minimax-m2.5, minimax/minimax-m2.5, minimaxai/minimax-m2.5, openai/gpt-oss-120b, openai/gpt-oss-20b, openai/gpt-oss-safeguard-20b, orpheus-v1-english, qwen/qwen3-vl-32b-instruct, whisper-large-v3LPU v2, Llama / Kimi / GPT-OSS, Llama 3.1 8B Instant, Llama 3.3 70B Versatile, Orpheus English, Qwen 3.6 27B, Whisper V3 Large, canopylabs/orpheus-arabic-saudi, canopylabs/orpheus-v1-english, kimi-k2-instruct-0905, meta-llama/llama-4-maverick-17b-128e-instruct, minimax-m2.5, minimax/minimax-m2.5, minimaxai/minimax-m2.5, openai/gpt-oss-120b, openai/gpt-oss-20b, openai/gpt-oss-safeguard-20b, orpheus-arabic-saudi, orpheus-v1-english, qwen/qwen3-vl-32b-instruct, qwen3-vl-32b-instruct, whisper-large-v3

    source
  • model

    Groq LPU added minimax-m2.5 to its model stack

    LPU v2, Llama / Kimi / GPT-OSS, Llama 3.1 8B Instant, Llama 3.3 70B Versatile, Orpheus English, Qwen 3.6 27B, Whisper V3 Large, canopylabs/orpheus-arabic-saudi, canopylabs/orpheus-v1-english, kimi-k2-instruct-0905, meta-llama/llama-4-maverick-17b-128e-instruct, minimax/minimax-m2.5, minimaxai/minimax-m2.5, openai/gpt-oss-120b, openai/gpt-oss-20b, openai/gpt-oss-safeguard-20b, orpheus-v1-english, qwen/qwen3-vl-32b-instruct, whisper-large-v3LPU v2, Llama / Kimi / GPT-OSS, Llama 3.1 8B Instant, Llama 3.3 70B Versatile, Orpheus English, Qwen 3.6 27B, Whisper V3 Large, canopylabs/orpheus-arabic-saudi, canopylabs/orpheus-v1-english, kimi-k2-instruct-0905, meta-llama/llama-4-maverick-17b-128e-instruct, minimax-m2.5, minimax/minimax-m2.5, minimaxai/minimax-m2.5, openai/gpt-oss-120b, openai/gpt-oss-20b, openai/gpt-oss-safeguard-20b, orpheus-v1-english, qwen/qwen3-vl-32b-instruct, whisper-large-v3

    source
  • capability

    New capabilities: Content Moderation

    24 tracked25 tracked · +Content Moderation

    source
  • version

    Groq LPU moved to LPU / GroqCloud API

    GroqCloudLPU / GroqCloud API

    source
  • model

    Groq LPU added minimax/minimax-m2.5 to its model stack

    GPT OSS 120B, GPT OSS 20B, LPU v2, Llama / Kimi / GPT-OSS, Llama 3.1 8B Instant, Llama 3.3 70B Versatile, Orpheus English, Qwen 3.6 27B, Whisper V3 Large, canopylabs/orpheus-arabic-saudi, canopylabs/orpheus-v1-english, gpt-oss-120b, gpt-oss-20b, gpt-oss-safeguard-20b, kimi-k2-instruct-0905, llama-3.1-8b-instant, llama-3.3-70b-versatile, meta-llama/llama-4-maverick-17b-128e-instruct, minimaxai/minimax-m2.5, openai/gpt-oss-120b, openai/gpt-oss-20b, openai/gpt-oss-safeguard-20b, orpheus-v1-english, qwen-3.6-27b, qwen/qwen3-vl-32b-instruct, whisper-large-v3GPT OSS 120B, GPT OSS 20B, LPU v2, Llama / Kimi / GPT-OSS, Llama 3.1 8B Instant, Llama 3.3 70B Versatile, Orpheus English, Qwen 3.6 27B, Whisper V3 Large, canopylabs/orpheus-arabic-saudi, canopylabs/orpheus-v1-english, gpt-oss-120b, gpt-oss-20b, gpt-oss-safeguard-20b, kimi-k2-instruct-0905, llama-3.1-8b-instant, llama-3.3-70b-versatile, meta-llama/llama-4-maverick-17b-128e-instruct, minimax/minimax-m2.5, minimaxai/minimax-m2.5, openai/gpt-oss-120b, openai/gpt-oss-20b, openai/gpt-oss-safeguard-20b, orpheus-v1-english, qwen-3.6-27b, qwen/qwen3-vl-32b-instruct, whisper-large-v3

    source
  • capability

    New capabilities: Text Generation

    239

    source
  • model

    Groq LPU added openai/gpt-oss-safeguard-20b, canopylabs/orpheus-v1-english to its model stack

    GPT OSS 120B, GPT OSS 20B, LPU v2, Llama / Kimi / GPT-OSS, Llama 3.1 8B Instant, Llama 3.3 70B Versatile, Orpheus English, Qwen 3.6 27B, Whisper V3 Large, canopylabs/orpheus-arabic-saudi, gpt-oss-120b, gpt-oss-20b, gpt-oss-safeguard-20b, kimi-k2-instruct-0905, llama-3.1-8b-instant, llama-3.3-70b-versatile, meta-llama/llama-4-maverick-17b-128e-instruct, minimaxai/minimax-m2.5, openai/gpt-oss-120b, openai/gpt-oss-20b, orpheus-v1-english, qwen-3.6-27b, qwen/qwen3-vl-32b-instruct, whisper-large-v3GPT OSS 120B, GPT OSS 20B, LPU v2, Llama / Kimi / GPT-OSS, Llama 3.1 8B Instant, Llama 3.3 70B Versatile, Orpheus English, Qwen 3.6 27B, Whisper V3 Large, canopylabs/orpheus-arabic-saudi, canopylabs/orpheus-v1-english, gpt-oss-120b, gpt-oss-20b, gpt-oss-safeguard-20b, kimi-k2-instruct-0905, llama-3.1-8b-instant, llama-3.3-70b-versatile, meta-llama/llama-4-maverick-17b-128e-instruct, minimaxai/minimax-m2.5, openai/gpt-oss-120b, openai/gpt-oss-20b, openai/gpt-oss-safeguard-20b, orpheus-v1-english, qwen-3.6-27b, qwen/qwen3-vl-32b-instruct, whisper-large-v3

    source
  • capability

    New capabilities: OpenAI API compatibility, Model Context Protocol (MCP) Connectors, Text to Speech (Orpheus)

    189

    source
  • model

    Groq LPU added openai/gpt-oss-120b, openai/gpt-oss-20b, meta-llama/llama-4-maverick-17b-128e-instruct, minimaxai/minimax-m2.5, qwen/qwen3-vl-32b-instruct, canopylabs/orpheus-arabic-saudi to its model stack

    GPT OSS 120B, GPT OSS 20B, LPU v2, Llama / Kimi / GPT-OSS, Llama 3.1 8B Instant, Llama 3.3 70B Versatile, Orpheus English, Qwen 3.6 27B, Whisper V3 Large, gpt-oss-120b, gpt-oss-20b, gpt-oss-safeguard-20b, kimi-k2-instruct-0905, llama-3.1-8b-instant, llama-3.3-70b-versatile, orpheus-v1-english, qwen-3.6-27b, whisper-large-v3GPT OSS 120B, GPT OSS 20B, LPU v2, Llama / Kimi / GPT-OSS, Llama 3.1 8B Instant, Llama 3.3 70B Versatile, Orpheus English, Qwen 3.6 27B, Whisper V3 Large, canopylabs/orpheus-arabic-saudi, gpt-oss-120b, gpt-oss-20b, gpt-oss-safeguard-20b, kimi-k2-instruct-0905, llama-3.1-8b-instant, llama-3.3-70b-versatile, meta-llama/llama-4-maverick-17b-128e-instruct, minimaxai/minimax-m2.5, openai/gpt-oss-120b, openai/gpt-oss-20b, orpheus-v1-english, qwen-3.6-27b, qwen/qwen3-vl-32b-instruct, whisper-large-v3

    source
  • capability

    New capabilities: Reasoning

    1710

    source
  • model

    Groq LPU added GPT OSS 20B, GPT OSS 120B, Llama 3.3 70B Versatile, Llama 3.1 8B Instant, Qwen 3.6 27B, Whisper V3 Large, Orpheus English to its model stack

    LPU v2, Llama / Kimi / GPT-OSS, gpt-oss-120b, gpt-oss-20b, gpt-oss-safeguard-20b, kimi-k2-instruct-0905, llama-3.1-8b-instant, llama-3.3-70b-versatile, orpheus-v1-english, qwen-3.6-27b, whisper-large-v3GPT OSS 120B, GPT OSS 20B, LPU v2, Llama / Kimi / GPT-OSS, Llama 3.1 8B Instant, Llama 3.3 70B Versatile, Orpheus English, Qwen 3.6 27B, Whisper V3 Large, gpt-oss-120b, gpt-oss-20b, gpt-oss-safeguard-20b, kimi-k2-instruct-0905, llama-3.1-8b-instant, llama-3.3-70b-versatile, orpheus-v1-english, qwen-3.6-27b, whisper-large-v3

    source
  • capability

    New capabilities: Fast LLM Inference, OpenAI-Compatible API, Structured Outputs

    139

    source
  • capability

    Context window now 131k

    —131k

    source
  • model

    Groq LPU added gpt-oss-20b, gpt-oss-120b, gpt-oss-safeguard-20b, llama-3.3-70b-versatile, llama-3.1-8b-instant, qwen-3.6-27b, kimi-k2-instruct-0905, orpheus-v1-english, whisper-large-v3 to its model stack

    LPU v2, Llama / Kimi / GPT-OSSLPU v2, Llama / Kimi / GPT-OSS, gpt-oss-120b, gpt-oss-20b, gpt-oss-safeguard-20b, kimi-k2-instruct-0905, llama-3.1-8b-instant, llama-3.3-70b-versatile, orpheus-v1-english, qwen-3.6-27b, whisper-large-v3

    source
  • capability

    New capabilities: Fast LLM inference, OpenAI Compatibility, Prompt Caching

    49

    source
  • pricing

    Token pricing cut across the open-model catalogue

    0.080.05

    source
  • capability

    New capabilities: Fast AI Inference, LPU AI Inference Engine

    3 tracked5 tracked · +Fast AI Inference, LPU AI Inference Engine

    source
  • model

    Groq LPU added Llama-2 70B, Mistral 7B to its model stack

    —Llama-2 70B, Mistral 7B

    source
  • capability

    New capabilities: LPU Inference Engine, Ultra-low Latency, Kernel-less Compiler

    4 tracked7 tracked · +LPU Inference Engine, Ultra-low Latency, Kernel-less Compiler

    source
  • version

    Groq LPU moved to LPU Inference Engine

    GroqChip / LPULPU Inference Engine

    source
  • capability

    New capabilities: deterministic execution, ultra low latency, compiler solutions

    6 tracked10 tracked · +deterministic execution, ultra low latency, compiler solutions, static profiling

    source
  • version

    Groq LPU moved to GroqChip / LPU

    GroqChipGroqChip / LPU

    source
  • capability

    New capabilities: Predictability, Velocity, Accuracy

    6 tracked11 tracked · +Predictability, Velocity, Accuracy, Scalability, Determinism

    source
  • model

    Groq LPU added GroqChip to its model stack

    —GroqChip

    source
  • capability

    New capabilities: Real-Time AI & HPC, Near-linear Scaling, Single-Core TSP Architecture

    6 tracked9 tracked · +Real-Time AI & HPC, Near-linear Scaling, Single-Core TSP Architecture

    source
  • version

    Groq LPU moved to GroqChip

    LPUGroqChip

    source
  • capability

    New capabilities: Low Latency, Deterministic Architecture, Predictable Execution

    5 tracked11 tracked · +Low Latency, Deterministic Architecture, Predictable Execution, RealScale Networking, TruePoint Technology, Static Profiling

    source
Subscribe to Groq LPU changes

Groq LPU compared

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

Groq LPU vs Tenstorrent BlackholeGroq LPU vs Google TPU IronwoodGroq LPU vs Cerebras WSE-3Groq LPU vs NVIDIA BlackwellGroq LPU vs AWS Trainium2Groq LPU 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/groq

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