Vertical Tabs
Get Started
Meta

Llama 3.2 1B Instruct

Model cardHugging FaceLaunch postWebsite
Llama 3.2 1B Instruct launch post cover
Context
60K
Released
2y ago
Parameters
1B
Input / 1M tokens
$0.03
Output / 1M tokens
$0.20

Sources:OpenRouter

Benchmarks

DeepSWE is the primary coding benchmark where an exact model evaluation exists. Other sourced evaluations appear below; missing scores are not zero.

Not evaluated in DeepSWE 1.1 as of 2026-09-22. Results are only shown for exact model versions in the official dataset. Official leaderboard.

Other published benchmarks

EvaluationReported score and source
GPQA diamond
23.9%
Epoch AI · Llama-3.2-1B-Instruct
Epoch AI evaluation
OTIS Mock AIME 2024-2025
0.6%
Epoch AI · Llama-3.2-1B-Instruct
Epoch AI evaluation

Benchmark compilation: Epoch AI, Capabilities & benchmarking (CC BY attribution). Scores are converted to percentages and model names matched to the Vertical Tabs catalog. Individual evaluations are credited to their original sources. External results may carry separate licences and evaluation configurations; do not interpret scores from different harnesses as directly comparable.

Specifications

Developer
Meta
Model family
Llama
Type
Language model
Predecessor
—
Successor
—
Released
September 25, 2024
Model ID
meta-llama/llama-3.2-1b-instruct
Openness
Open weights
License
—
Parameters
1B
Input modalities
text
Output modalities
text
Modality
text->text
Native context
128K
Tokenizer
Llama3
Max output tokens
54K
Knowledge cutoff
2023-12-31
Hugging Face
meta-llama/Llama-3.2-1B-Instruct
Source freshness
Endpoints synced Oct 15, 2026, 12:31 PM UTC

Open-weight builds

Ollama variants

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

60 builds
VariantParametersTypeQuantizationSizeContextInput
llama3.2:1b-instruct-fp161BinstructFP162.5 GB128Ktext
llama3.2:1b-instruct-q2_K1BinstructQ2_K581 MB128Ktext
llama3.2:1b-instruct-q3_K_L1BinstructQ3_K_L733 MB128Ktext
llama3.2:1b-instruct-q3_K_M1BinstructQ3_K_M691 MB128Ktext
llama3.2:1b-instruct-q3_K_S1BinstructQ3_K_S642 MB128Ktext
llama3.2:1b-instruct-q4_01BinstructQ4_0771 MB128Ktext
llama3.2:1b-instruct-q4_11BinstructQ4_1832 MB128Ktext
llama3.2:1b-instruct-q4_K_M1BinstructQ4_K_M808 MB128Ktext
llama3.2:1b-instruct-q4_K_S1BinstructQ4_K_S776 MB128Ktext
llama3.2:1b-instruct-q5_01BinstructQ5_0893 MB128Ktext
llama3.2:1b-instruct-q5_11BinstructQ5_1953 MB128Ktext
llama3.2:1b-instruct-q5_K_M1BinstructQ5_K_M912 MB128Ktext
llama3.2:1b-instruct-q5_K_S1BinstructQ5_K_S893 MB128Ktext
llama3.2:1b-instruct-q6_K1BinstructQ6_K1 GB128Ktext
llama3.2:1b1BinstructQ8_01.3 GB128Ktext
llama3.2:1b-text-fp161BtextFP162.5 GB128Ktext
llama3.2:1b-text-q2_K1BtextQ2_K581 MB128Ktext
llama3.2:1b-text-q3_K_L1BtextQ3_K_L733 MB128Ktext
llama3.2:1b-text-q3_K_M1BtextQ3_K_M691 MB128Ktext
llama3.2:1b-text-q3_K_S1BtextQ3_K_S642 MB128Ktext
llama3.2:1b-text-q4_01BtextQ4_0771 MB128Ktext
llama3.2:1b-text-q4_11BtextQ4_1832 MB128Ktext
llama3.2:1b-text-q4_K_M1BtextQ4_K_M808 MB128Ktext
llama3.2:1b-text-q4_K_S1BtextQ4_K_S776 MB128Ktext
llama3.2:1b-text-q5_01BtextQ5_0893 MB128Ktext
llama3.2:1b-text-q5_11BtextQ5_1953 MB128Ktext
llama3.2:1b-text-q5_K_M1BtextQ5_K_M912 MB128Ktext
llama3.2:1b-text-q5_K_S1BtextQ5_K_S893 MB128Ktext
llama3.2:1b-text-q6_K1BtextQ6_K1 GB128Ktext
llama3.2:1b-text-q8_01BtextQ8_01.3 GB128Ktext
llama3.2:3b-instruct-fp163BinstructFP166.4 GB128Ktext
llama3.2:3b-instruct-q2_K3BinstructQ2_K1.4 GB128Ktext
llama3.2:3b-instruct-q3_K_L3BinstructQ3_K_L1.8 GB128Ktext
llama3.2:3b-instruct-q3_K_M3BinstructQ3_K_M1.7 GB128Ktext
llama3.2:3b-instruct-q3_K_S3BinstructQ3_K_S1.5 GB128Ktext
llama3.2:3b-instruct-q4_03BinstructQ4_01.9 GB128Ktext
llama3.2:3b-instruct-q4_13BinstructQ4_12.1 GB128Ktext
llama3.2:latest3BinstructQ4_K_M2 GB128Ktext
llama3.2:3b-instruct-q4_K_S3BinstructQ4_K_S1.9 GB128Ktext
llama3.2:3b-instruct-q5_03BinstructQ5_02.3 GB128Ktext
llama3.2:3b-instruct-q5_13BinstructQ5_12.4 GB128Ktext
llama3.2:3b-instruct-q5_K_M3BinstructQ5_K_M2.3 GB128Ktext
llama3.2:3b-instruct-q5_K_S3BinstructQ5_K_S2.3 GB128Ktext
llama3.2:3b-instruct-q6_K3BinstructQ6_K2.6 GB128Ktext
llama3.2:3b-instruct-q8_03BinstructQ8_03.4 GB128Ktext
llama3.2:3b-text-fp163BtextFP166.4 GB128Ktext
llama3.2:3b-text-q2_K3BtextQ2_K1.4 GB128Ktext
llama3.2:3b-text-q3_K_L3BtextQ3_K_L1.8 GB128Ktext
llama3.2:3b-text-q3_K_M3BtextQ3_K_M1.7 GB128Ktext
llama3.2:3b-text-q3_K_S3BtextQ3_K_S1.5 GB128Ktext
llama3.2:3b-text-q4_03BtextQ4_01.9 GB128Ktext
llama3.2:3b-text-q4_13BtextQ4_12.1 GB128Ktext
llama3.2:3b-text-q4_K_M3BtextQ4_K_M2 GB128Ktext
llama3.2:3b-text-q4_K_S3BtextQ4_K_S1.9 GB128Ktext
llama3.2:3b-text-q5_03BtextQ5_02.3 GB128Ktext
llama3.2:3b-text-q5_13BtextQ5_12.4 GB128Ktext
llama3.2:3b-text-q5_K_M3BtextQ5_K_M2.3 GB128Ktext
llama3.2:3b-text-q5_K_S3BtextQ5_K_S2.3 GB128Ktext
llama3.2:3b-text-q6_K3BtextQ6_K2.6 GB128Ktext
llama3.2:3b-text-q8_03BtextQ8_03.4 GB128Ktext

Infrastructure

Provider availability

All providers
ProviderEndpointContextInput /M tokensOutput /M tokensCached input /M tokensCache write /M tokensUptime 1dData handling
Cloudflarecloudflare60K$0.03$0.20——100.00%
ZDR: No

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

ZDR source Privacy policy

Checked 2026-10-15

Provider data:OpenRouter

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