FitMyLLM
Reka AI/Dense

RReka Flash 3

Reka Flash 3 is a 21B general-purpose reasoning model that was trained from scratch. It was trained in synthetic and public datasets for supervised finetuning, followed by RLOO with model-based and rule-based rewards.

chatreasoningcoding
21B
Parameters
32K
Context length
11
Benchmarks
10
Quantizations
0
Architecture
Dense
Released
2025-03-15
Layers
44
KV Heads
8
Head Dim
96
Family
other

Quantization Options

Context length:
QuantBitsVRAM @ 16KQuality
Q3_K_M4
12.5 GB
11.0 + 1.5 KV
low
Q3_K_L4.3
13.3 GB
11.8 + 1.5 KV
moderate
IQ4_XS4.46
13.7 GB
12.2 + 1.5 KV
moderate
Q4_K_S4.67
14.3 GB
12.7 + 1.5 KV
moderate
Q4_K_M4.89
14.9 GB
13.3 + 1.5 KV
good
Q5_K_S5.57
16.7 GB
15.1 + 1.5 KV
good
Q5_K_M5.7
17.0 GB
15.5 + 1.5 KV
good
Q6_K6.56
19.3 GB
17.7 + 1.5 KV
excellent
Q8_08.5
24.3 GB
22.8 + 1.5 KV
lossless
FP1616
44.0 GB
42.5 + 1.5 KV
lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

Deploying for a team or in production? Size GPUs, cost & scaling in Enterprise →
▸ READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN REKA FLASH 3 NOW

Rent on Vast.ai →Or RunPod →

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Benchmarks (11)

MATH-50089.3
MMLU-PRO65.0
GPQA Diamond52.9
LiveCodeBench43.5
AIME33.7
AA Math33.7
IFBench30.4
SciCode26.7
AA Intelligence9.5
AA Coding8.9
HLE5.1

Run this model

▸Easiest way to get started·Beginners
DOCS ↗
curl -fsSL https://ollama.com/install.sh | sh
$ollama run other:21b-q4_K_M

Tag may need adjustment — check ollama.com/library/other for available tags.

▸ SETUP GUIDE
>_

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more
Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines
HuggingFaceGGUF DownloadsBuild Hardware

GPUs that can run this model

At Q4_K_M quantization. Sorted by minimum VRAM.

NVIDIA RTX 5080
16 GB VRAM • 960 GB/s
NVIDIA
$999
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA RTX 5070 Ti
16 GB VRAM • 896 GB/s
NVIDIA
$749
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA RTX 4080 SUPER
16 GB VRAM • 736 GB/s
NVIDIA
$999
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA RTX 4080
16 GB VRAM • 717 GB/s
NVIDIA
$1199
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA RTX 4070 Ti SUPER
16 GB VRAM • 672 GB/s
NVIDIA
$799
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA RTX 4060 Ti 16GB
16 GB VRAM • 288 GB/s
NVIDIA
$499
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
AMD RX 7900 GRE
16 GB VRAM • 576 GB/s
AMD
$549
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
AMD RX 7800 XT
16 GB VRAM • 624 GB/s
AMD
$499
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
AMD RX 7600 XT
16 GB VRAM • 288 GB/s
AMD
$329
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
AMD RX 6950 XT
16 GB VRAM • 576 GB/s
AMD
$449
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
AMD RX 6900 XT
16 GB VRAM • 512 GB/s
AMD
$469
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
AMD RX 6800 XT
16 GB VRAM • 512 GB/s
AMD
$599
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
AMD RX 6800
16 GB VRAM • 512 GB/s
AMD
$599
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
Intel Arc A770 16GB
16 GB VRAM • 560 GB/s
INTEL
$349
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
Apple M1 Pro (16GB)
16 GB VRAM • 200 GB/s
APPLE
$999
Apple M2 Pro (16GB)
16 GB VRAM • 200 GB/s
APPLE
$1299
Apple M4 (16GB)
16 GB VRAM • 120 GB/s
APPLE
$499
NVIDIA Tesla T4 16GB
16 GB VRAM • 320 GB/s
NVIDIA
$800
NVIDIA V100 PCIe 16GB
16 GB VRAM • 900 GB/s
NVIDIA
$2000
AMD RX 9070 XT
16 GB VRAM • 640 GB/s
AMD
$599
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
AMD RX 9070
16 GB VRAM • 672 GB/s
AMD
$549
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
Apple M1 (16GB)
16 GB VRAM • 68.25 GB/s
APPLE
$699
Apple M2 (16GB)
16 GB VRAM • 100 GB/s
APPLE
$799
Apple M3 (16GB)
16 GB VRAM • 100 GB/s
APPLE
$799
NVIDIA RTX 5060 Ti 16GB
16 GB VRAM • 448 GB/s
NVIDIA
$429
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA Tesla P100 DGXS
16 GB VRAM • 732 GB/s
NVIDIA
NVIDIA Tesla P100 PCIe 16 GB
16 GB VRAM • 732 GB/s
NVIDIA
NVIDIA Tesla P100 SXM2
16 GB VRAM • 732 GB/s
NVIDIA
NVIDIA Tesla V100 PCIe 16 GB
16 GB VRAM • 897 GB/s
NVIDIA
NVIDIA Tesla V100 SXM2 16 GB
16 GB VRAM • 1130 GB/s
NVIDIA

Find the best GPU for Reka Flash 3

Build Hardware for Reka Flash 3
▸ SPEC SHEET

Reka Flash 3 — 21B Dense.

▸ SPECIFICATIONS
PARAMETERS
21B
ARCHITECTURE
Dense Transformer
CONTEXT LENGTH
32K tokens
CAPABILITIES
chat, reasoning, coding
RELEASE DATE
2025-03-15
PROVIDER
Reka AI
FAMILY
other
▸ VRAM REQUIREMENTS
QUANTBPWVRAMQUALITY
Q3_K_M411.0 GB88%
Q3_K_L4.311.8 GB90%
IQ4_XS4.4612.2 GB92%
Q4_K_S4.6712.7 GB93%
Q4_K_M4.8913.3 GB94%
Q5_K_S5.5715.1 GB96%
Q5_K_M5.715.5 GB96%
Q6_K6.5617.7 GB97%
Q8_08.522.8 GB100%
FP161642.5 GB100%
§ 01BENCHMARK SCORES
MMLU-PRO65.0
GPQA Diamond52.9
LiveCodeBench43.5
AIME33.7
HLE5.1
AA Intelligence9.5
AA Coding8.9
AA Math33.7
aa_ifbench30.4
aa_scicode26.7
MATH-50089.3
§ 03COMPATIBLE GPUs
30 @ Q4_K_M
NVIDIA RTX 5080
16 GB · 960 GB/s
NVIDIA RTX 5070 Ti
16 GB · 896 GB/s
NVIDIA RTX 4080 SUPER
16 GB · 736 GB/s
NVIDIA RTX 4080
16 GB · 717 GB/s
NVIDIA RTX 4070 Ti SUPER
16 GB · 672 GB/s
NVIDIA RTX 4060 Ti 16GB
16 GB · 288 GB/s
AMD RX 7900 GRE
16 GB · 576 GB/s
AMD RX 7800 XT
16 GB · 624 GB/s
AMD RX 7600 XT
16 GB · 288 GB/s
AMD RX 6950 XT
16 GB · 576 GB/s
AMD RX 6900 XT
16 GB · 512 GB/s
AMD RX 6800 XT
16 GB · 512 GB/s
AMD RX 6800
16 GB · 512 GB/s
Intel Arc A770 16GB
16 GB · 560 GB/s
Apple M1 Pro (16GB)
16 GB · 200 GB/s