FitMyLLM
DeepCogito/Mixture of Experts

DCogito v2 671B MoE

The Cogito v2 LLMs are instruction tuned generative models. All models are released under an open license for commercial use.

chatreasoningtool_useagentic
671B
Parameters (37B active)
160K
Context length
9
Benchmarks
17
Quantizations
0
Architecture
MoE
Released
2025-09-10
Layers
61
KV Heads
128
Head Dim
64
Family
cogito

Quantization Options

QuantBitsVRAMQuality
IQ2_XXS2.38200.1 GBlow
IQ2_M2.93246.2 GBlow
Q2_K3.16265.5 GBlow
IQ3_XXS3.25273.1 GBlow
IQ3_XS3.5294.1 GBlow
Q3_K_S3.64305.8 GBlow
IQ3_M3.76315.9 GBlow
Q3_K_M4336.0 GBlow
Q3_K_L4.3361.2 GBmoderate
IQ4_XS4.46374.6 GBmoderate
Q4_K_S4.67392.2 GBmoderate
Q4_K_M4.89410.6 GBgood
Q5_K_S5.57467.7 GBgood
Q5_K_M5.7478.6 GBgood
Q6_K6.56550.7 GBexcellent
Q8_08.5713.4 GBlossless
FP16161342.5 GBlossless

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

▸ READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN COGITO V2 671B MOE NOW

Rent on RunPod →Or Vast.ai →

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

Community Ratings

Loading ratings...

Benchmarks (9)

MMLU-PRO84.9
GPQA Diamond76.8
AIME72.7
LiveCodeBench68.8
IFBench46.3
SciCode41.0
AA Long Context21.7
Terminal-Bench16.7
HLE11.0

Run this model

▸Easiest way to get started·Beginners
DOCS ↗
curl -fsSL https://ollama.com/install.sh | sh
$ollama run cogito:671b-q4_K_M

Tag may need adjustment — check ollama.com/library/cogito 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

Find the best GPU for Cogito v2 671B MoE

Build Hardware for Cogito v2 671B MoE
▸ SPEC SHEET

Cogito v2 671B MoE — 671B MoE.

▸ SPECIFICATIONS
PARAMETERS
671B (37B active)
ARCHITECTURE
Mixture of Experts
CONTEXT LENGTH
160K tokens
CAPABILITIES
chat, reasoning, tool_use, agentic
RELEASE DATE
2025-09-10
PROVIDER
DeepCogito
FAMILY
cogito
▸ VRAM REQUIREMENTS
QUANTBPWVRAMQUALITY
IQ2_XXS2.38200.1 GB65%
IQ2_M2.93246.2 GB75%
Q2_K3.16265.5 GB78%
IQ3_XXS3.25273.1 GB82%
IQ3_XS3.5294.1 GB84%
Q3_K_S3.64305.8 GB85%
IQ3_M3.76315.9 GB86%
Q3_K_M4336.0 GB88%
Q3_K_L4.3361.2 GB90%
IQ4_XS4.46374.6 GB92%
Q4_K_S4.67392.2 GB93%
Q4_K_M4.89410.6 GB94%
Q5_K_S5.57467.7 GB96%
Q5_K_M5.7478.6 GB96%
Q6_K6.56550.7 GB97%
Q8_08.5713.4 GB100%
FP16161342.5 GB100%
§ 01BENCHMARK SCORES
MMLU-PRO84.9
GPQA Diamond76.8
LiveCodeBench68.8
AIME72.7
HLE11.0
aa_ifbench46.3
aa_terminal_bench16.7
aa_scicode41.0
aa_lcr21.7