Meta, released Sep 25, 2024

Llama 3.2 1B Instructprice, context, benchmarks and release details

Provisional: not enough results to rank yet Open weights 93% confidence, 2 of 4 expected sources in

35.1

SI Score

#119 of 125 ranked models

Input, per 1M tokens: not yet reported
Output, per 1M tokens: not yet reported
Context window: 131K
Max output: 32K
Released: Sep 25, 2024

How this score is built

Coding —

Math 0.6

Preference 32.6

Reasoning 23.9

How Llama 3.2 1B Instruct's SI Score is calculated
Pillar	Score	Weight	Adds
Math	0.6	25% of 15%	0.1
Preference	32.6	25% of 15%	8.1
Reasoning	23.9	50% of 30%	12.0
Coding	Not measured yet, left out
Average of measured pillars	20.2
Evidence check	14.9 (3.0 of 6suites needed)
SI Score	35.1

Measured suites (3)

Related tasks and editions share one suite weight in each pillar, and one reliability-weighted breadth contribution across the model.

aime: 1.00 breadth; Math 0.6 × 0.50. 1 result ID: otis-mock-aime-2024-2025
gpqa: 1.00 breadth; Reasoning 23.9 × 0.50. 1 result ID: gpqa-diamond
lmarena-text: 1.00 breadth; Preference 32.6 × 1.00. 1 result ID: lmarena-text

Missing pillars are left out and the remaining weights are rescaled, so nothing counts as a zero. With fewer than 6 weighted suites, the average is pulled toward 50 until more results arrive. Still to report for this model, and not counted against it: Official model cards via models.dev, LiveBench. Full method

Method si-v5-suite-evidence-1, computed Oct 9, 2026, 18:16 UTC.

Around it on the leaderboard

117 Llama-3.1-8B-Instruct 36.3
118 GPT-4.1 mini 35.4
119 Llama 3.2 1B Instruct 35.1
120 Llama-3.3-70B-Instruct 34.6
121 GPT-4o (2024-08-06) 34.4

Full leaderboard

Benchmark results

3 benchmarks, 3 results

Each row shows the best published result. Where a model was tested at several settings, such as reasoning effort, open the row to see each one. Hover or tap a value for its source.

GPQA Diamondreasoning 23.9% 23.9

OTIS Mock AIME 2024–2025math 0.6% 0.6

LMArena Textpreference 1054.6 elo 32.6

Normalization uses a fixed 0–100 scale for each unit, independent of other models. Compare evaluation conditions before reading a small gap as decisive. “Lab-reported” marks the provider's own published figure.

Details and sources

Open weights: Yes
License: Llama 3.2 Community License
Input modalities: text
First seen by SuperIndex: Oct 9, 2026
Coverage: 75% of expected source weight

Reported (2)

Epoch AI BenchmarkingOct 9, 2026
LMArena / ArenaOct 9, 2026

Awaiting (2)

Official model cards via models.dev7% of weight
LiveBench19% of weight

Confidence rises as pending sources publish. Some sources never cover some models, so confidence reaches 100% at 80% of expected weight.