Meta, open weights
Llama 3.2 3B
Llama 3.2 3B by Meta ranks 321st of 354 ranked models on the Noometry Index as of October 2026, with a score of 28.9. Its strongest category is agentic & tool use, where it ranks 143rd. API pricing starts at $0.05 per million input tokens and $0.33 per million output tokens, with a 131K-token context window.
Last verified
Specifications
- Noometry rank
- #321 of 354
- Index score
- 28.9
- Evidence
- Confirmed 18 results
- Provider
Meta- Released
- September 24, 2024
- Weights
- Open weights
- Reasoning
- Unknown
- Context window
- 131K
- Max output
- 118K
- Input price
- $0.05 / M
- Output price
- $0.33 / M
- Blended price
- $0.12 / M
- Output speed
- Not measured
- Value
- #33 of 219
- Knowledge cutoff
- Unknown
- Input
- text
- Hugging Face
- meta-llama/Llama-3.2-3B-Instruct
Category scores
Each category score combines every public result we have in that category.
- Coding 27.6
- Agentic & Tool Use 20.1
- Reasoning 21.0
- Math 32.4
- Knowledge 29.7
- Multilingual 26.2
- Instruction Following 56.0
- Long Context 33.4
- Writing & Preference 24.7
| Category | Score | Rank | Results |
|---|---|---|---|
| Coding | 27.6 | #319 | 3 |
| Agentic & Tool Use | 20.1 | #143 | 2 |
| Reasoning | 21.0 | #228 | 1 |
| Math | 32.4 | #214 | 1 |
| Knowledge | 29.7 | #235 | 1 |
| Multilingual | 26.2 | #281 | 1 |
| Instruction Following | 56.0 | #275 | 1 |
| Long Context | 33.4 | #261 | 1 |
| Writing & Preference | 24.7 | #307 | 4 |
Strengths and weaknesses
Categories where Llama 3.2 3B places highest and lowest among the models ranked in each, with its score against that category's median.
Strongest categories
Weakest categories
| Category | Score | vs median | Rank |
|---|---|---|---|
| Writing & Preference | 24.7 | −29.1 | #307 of 312, top 99% |
| Multilingual | 26.2 | −21.2 | #281 of 297, top 95% |
| Coding | 27.6 | −11.1 | #319 of 340, top 94% |
Closest competitors
The models ranked just above and below Llama 3.2 3B. When scores are this close, price and speed are often the better way to choose.
| Model | Rank | Score | Blended $/M | Speed | |
|---|---|---|---|---|---|
| Llama 2-7B | #317 | 29.1 | — | — | Compare |
| Granite 4.0 Micro | #318 | 29.0 | $0.0408 | — | Compare |
| Claude 3 Sonnet | #319 | 29.0 | — | — | Compare |
| Qwen2.5 7B Instruct | #320 | 29.0 | $0.31 | — | Compare |
| Qwen1.5 4b Chat | #322 | 28.8 | — | — | Compare |
| Llama 3-70B | #323 | 28.8 | — | 104 | Compare |
| GPT-4o | #324 | 28.6 | $4.38 | — | Compare |
| Ministral 8B | #325 | 28.2 | $0.15 | — | Compare |
Sponsored placements are available on pages like this one. Advertise on Noometry
Benchmark results
Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.
Coding
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| BigCodeBench Instruct | 23.4% | #59 of 64, top 93% | BigCodeBench | 2024-09-25 | |
| LMArena Coding | 1098 | #270 of 294, top 92% | LMArena | 2026-10-08 | |
| BigCodeBench Complete | 28.3% | #61 of 66, top 93% | BigCodeBench | 2024-09-25 |
Agentic & Tool Use
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| Berkeley Function Calling Leaderboard | 21.9% | #44 of 49, top 90% | fc | Berkeley Function Calling Leaderboard | |
| BALROG | 10.1% | #33 of 35, top 95% | Epoch AI |
Reasoning
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Hard Prompts | 1095 | #269 of 297, top 91% | LMArena | 2026-10-08 |
Math
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Math | 1126 | #259 of 285, top 91% | LMArena | 2026-10-08 |
Knowledge
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Expert | 1090 | #252 of 273, top 93% | LMArena | 2026-10-08 |
Multilingual
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Non-English | 1019 | #281 of 297, top 95% | LMArena | 2026-10-08 | |
| LMArena Chinese | 1017 | #273 of 285, top 96% | LMArena | 2026-10-08 | |
| LMArena German | 1056 | #218 of 231, top 95% | LMArena | 2026-10-08 | |
| LMArena Russian | 949 | #280 of 283, top 99% | LMArena | 2026-10-08 |
Instruction Following
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Instruction Following | 1089 | #270 of 298, top 91% | LMArena | 2026-10-08 |
Long Context
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Longer Query | 1100 | #267 of 291, top 92% | LMArena | 2026-10-08 |
Writing & Preference
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Text | 1110 | #271 of 297, top 92% | LMArena | 2026-10-08 | |
| LMArena Creative Writing | 1094 | #265 of 295, top 90% | LMArena | 2026-10-08 | |
| EQ-Bench Creative Writing | 595 | #113 of 115, top 99% | EQ-Bench | ||
| LMArena Multi-Turn | 1105 | #266 of 295, top 91% | LMArena | 2026-10-08 |
API pricing by provider
| Route | Input $/M | Output $/M | Cached input $/M | Checked |
|---|---|---|---|---|
| openrouter | $0.05 | $0.33 | — | 2026-10-10 |
Compare Llama 3.2 3B
- Llama 3.2 3B vs Llama 3.1-405B
- Llama 3.2 3B vs Qwen2.5 7B Instruct
- Llama 3.2 3B vs Qwen1.5 4b Chat
- Llama 3.2 3B vs Claude 3 Sonnet
- Llama 3.2 3B vs Llama 3-70B
- Llama 3.2 3B vs Granite 4.0 Micro
- Llama 3.2 3B vs GPT-4o
- Llama 3.2 3B vs GPT-6 Astra
- Llama 3.2 3B vs Claude Fable 5.1
- Llama 3.2 3B vs Gemini 3.8 Flash
- Llama 3.2 3B vs Kimi K3
- Llama 3.2 3B vs Grok 4.6
- Llama 3.2 3B vs Qwen3.8 Max
- Llama 3.2 3B vs GLM-5.3
Other Meta models
- Muse Spark 1.354.8
- Muse Spark50.6
- Muse Spark 1.250.3
- Muse Spark 1.149.9
- Muse Glimmer41.7
- Codellama 70b Instruct33.7
- Llama 4 Maverick30.9
- Codellama 34b Instruct30.8
Frequently asked questions
How good is Llama 3.2 3B?
Llama 3.2 3B by Meta ranks 321st of 354 ranked models on the Noometry Index as of October 2026, with a score of 28.9. Its strongest category is agentic & tool use, where it ranks 143rd. API pricing starts at $0.05 per million input tokens and $0.33 per million output tokens, with a 131K-token context window.
How much does Llama 3.2 3B cost?
Llama 3.2 3B costs $0.05 per million input tokens and $0.33 per million output tokens on openrouter.
What is Llama 3.2 3B's context window?
Llama 3.2 3B accepts up to 131K tokens of input and can write up to 118K tokens in one response.
Is Llama 3.2 3B open source?
Yes. Llama 3.2 3B's weights are downloadable from Hugging Face (meta-llama/Llama-3.2-3B-Instruct); check the license for commercial terms.
What are Llama 3.2 3B's strengths and weaknesses?
Relative to other ranked models, Llama 3.2 3B places best in reasoning, math, knowledge and lowest in writing & preference, multilingual, coding.
What is Llama 3.2 3B best at?
Its best category is agentic & tool use, where it ranks 143rd on Noometry.