Sign Up

Llama 3.1 8B Instruct

Meta
Announcement
Llama 3.1 8B Instruct launch post cover

Cost/1M tokens

input
$0.02
output
$0.04
Context
131Ktokens
Released
2y ago
Size
8Bparameters

Open-weight builds

Ollama variants

Every published Ollama tag matched to this model, including parameter editions, instruction/base variants, and quantized builds.

89 builds
VariantParametersTypeQuantizationSizeContextInput
llama3.1:8b-instruct-fp168BinstructFP1616 GB128Ktext
llama3.1:8b-instruct-q2_K8BinstructQ2_K3.2 GB128Ktext
llama3.1:8b-instruct-q3_K_L8BinstructQ3_K_L4.3 GB128Ktext
llama3.1:8b-instruct-q3_K_M8BinstructQ3_K_M4 GB128Ktext
llama3.1:8b-instruct-q3_K_S8BinstructQ3_K_S3.7 GB128Ktext
llama3.1:8b-instruct-q4_08BinstructQ4_04.7 GB128Ktext
llama3.1:8b-instruct-q4_18BinstructQ4_15.1 GB128Ktext
llama3.1:latest8BinstructQ4_K_M4.9 GB128Ktext
llama3.1:8b-instruct-q4_K_S8BinstructQ4_K_S4.7 GB128Ktext
llama3.1:8b-instruct-q5_08BinstructQ5_05.6 GB128Ktext
llama3.1:8b-instruct-q5_18BinstructQ5_16.1 GB128Ktext
llama3.1:8b-instruct-q5_K_M8BinstructQ5_K_M5.7 GB128Ktext
llama3.1:8b-instruct-q5_K_S8BinstructQ5_K_S5.6 GB128Ktext
llama3.1:8b-instruct-q6_K8BinstructQ6_K6.6 GB128Ktext
llama3.1:8b-instruct-q8_08BinstructQ8_08.5 GB128Ktext
llama3.1:8b-text-fp168BtextFP1616 GB128Ktext
llama3.1:8b-text-q2_K8BtextQ2_K3.2 GB128Ktext
llama3.1:8b-text-q3_K_L8BtextQ3_K_L4.3 GB128Ktext
llama3.1:8b-text-q3_K_M8BtextQ3_K_M4 GB128Ktext
llama3.1:8b-text-q3_K_S8BtextQ3_K_S3.7 GB128Ktext
llama3.1:8b-text-q4_08BtextQ4_04.7 GB128Ktext
llama3.1:8b-text-q4_18BtextQ4_15.1 GB128Ktext
llama3.1:8b-text-q4_K_M8BtextQ4_K_M4.9 GB128Ktext
llama3.1:8b-text-q4_K_S8BtextQ4_K_S4.7 GB128Ktext
llama3.1:8b-text-q5_08BtextQ5_05.6 GB128Ktext
llama3.1:8b-text-q5_18BtextQ5_16.1 GB128Ktext
llama3.1:8b-text-q5_K_M8BtextQ5_K_M5.7 GB128Ktext
llama3.1:8b-text-q5_K_S8BtextQ5_K_S5.6 GB128Ktext
llama3.1:8b-text-q6_K8BtextQ6_K6.6 GB128Ktext
llama3.1:8b-text-q8_08BtextQ8_08.5 GB128Ktext
llama3.1:70b-instruct-fp1670BinstructFP16141 GB128Ktext
llama3.1:70b-instruct-q2_K70BinstructQ2_K26 GB128Ktext
llama3.1:70b-instruct-q3_K_L70BinstructQ3_K_L37 GB128Ktext
llama3.1:70b-instruct-q3_K_M70BinstructQ3_K_M34 GB128Ktext
llama3.1:70b-instruct-q3_K_S70BinstructQ3_K_S31 GB128Ktext
llama3.1:70b-instruct-q4_070BinstructQ4_040 GB128Ktext
llama3.1:70b70BinstructQ4_K_M43 GB128Ktext
llama3.1:70b-instruct-q4_K_S70BinstructQ4_K_S40 GB128Ktext
llama3.1:70b-instruct-q5_070BinstructQ5_049 GB128Ktext
llama3.1:70b-instruct-q5_170BinstructQ5_153 GB128Ktext
llama3.1:70b-instruct-q5_K_M70BinstructQ5_K_M50 GB128Ktext
llama3.1:70b-instruct-q5_K_S70BinstructQ5_K_S49 GB128Ktext
llama3.1:70b-instruct-q6_K70BinstructQ6_K58 GB128Ktext
llama3.1:70b-instruct-q8_070BinstructQ8_075 GB128Ktext
llama3.1:70b-text-fp1670BtextFP16141 GB128Ktext
llama3.1:70b-text-q2_K70BtextQ2_K26 GB128Ktext
llama3.1:70b-text-q3_K_L70BtextQ3_K_L37 GB128Ktext
llama3.1:70b-text-q3_K_M70BtextQ3_K_M34 GB128Ktext
llama3.1:70b-text-q3_K_S70BtextQ3_K_S31 GB128Ktext
llama3.1:70b-text-q4_070BtextQ4_040 GB128Ktext
llama3.1:70b-text-q4_170BtextQ4_144 GB128Ktext
llama3.1:70b-text-q4_K_M70BtextQ4_K_M43 GB128Ktext
llama3.1:70b-text-q4_K_S70BtextQ4_K_S40 GB128Ktext
llama3.1:70b-text-q5_070BtextQ5_049 GB128Ktext
llama3.1:70b-text-q5_170BtextQ5_153 GB128Ktext
llama3.1:70b-text-q5_K_M70BtextQ5_K_M50 GB128Ktext
llama3.1:70b-text-q5_K_S70BtextQ5_K_S49 GB128Ktext
llama3.1:70b-text-q6_K70BtextQ6_K58 GB128Ktext
llama3.1:70b-text-q8_070BtextQ8_075 GB128Ktext
llama3.1:405b-instruct-fp16405BinstructFP16812 GB128Ktext
llama3.1:405b-instruct-q2_K405BinstructQ2_K149 GB128Ktext
llama3.1:405b-instruct-q3_K_L405BinstructQ3_K_L213 GB128Ktext
llama3.1:405b-instruct-q3_K_M405BinstructQ3_K_M195 GB128Ktext
llama3.1:405b-instruct-q3_K_S405BinstructQ3_K_S175 GB128Ktext
llama3.1:405b-instruct-q4_0405BinstructQ4_0229 GB128Ktext
llama3.1:405b-instruct-q4_1405BinstructQ4_1254 GB128Ktext
llama3.1:405b405BinstructQ4_K_M243 GB128Ktext
llama3.1:405b-instruct-q4_K_S405BinstructQ4_K_S231 GB128Ktext
llama3.1:405b-instruct-q5_0405BinstructQ5_0279 GB128Ktext
llama3.1:405b-instruct-q5_1405BinstructQ5_1305 GB128Ktext
llama3.1:405b-instruct-q5_K_M405BinstructQ5_K_M287 GB128Ktext
llama3.1:405b-instruct-q5_K_S405BinstructQ5_K_S279 GB128Ktext
llama3.1:405b-instruct-q6_K405BinstructQ6_K333 GB128Ktext
llama3.1:405b-instruct-q8_0405BinstructQ8_0431 GB128Ktext
llama3.1:405b-text-fp16405BtextFP16812 GB128Ktext
llama3.1:405b-text-q2_K405BtextQ2_K149 GB128Ktext
llama3.1:405b-text-q3_K_L405BtextQ3_K_L213 GB128Ktext
llama3.1:405b-text-q3_K_M405BtextQ3_K_M195 GB128Ktext
llama3.1:405b-text-q3_K_S405BtextQ3_K_S175 GB128Ktext
llama3.1:405b-text-q4_0405BtextQ4_0229 GB128Ktext
llama3.1:405b-text-q4_1405BtextQ4_1254 GB128Ktext
llama3.1:405b-text-q4_K_M405BtextQ4_K_M243 GB128Ktext
llama3.1:405b-text-q4_K_S405BtextQ4_K_S231 GB128Ktext
llama3.1:405b-text-q5_0405BtextQ5_0279 GB128Ktext
llama3.1:405b-text-q5_1405BtextQ5_1305 GB128Ktext
llama3.1:405b-text-q5_K_M405BtextQ5_K_M287 GB128Ktext
llama3.1:405b-text-q5_K_S405BtextQ5_K_S279 GB128Ktext
llama3.1:405b-text-q6_K405BtextQ6_K333 GB128Ktext
llama3.1:405b-text-q8_0405BtextQ8_0431 GB128Ktext

Deployment graph

Can I run this?

Vertical Tabs only shows compatibility when it has structured evidence tying this exact model to hardware, runtime, quantization, memory use, and observed performance.

Browse hardware

To give you the best experience, we use cookies for personalization, insights, and performance. Learn more.