NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/Llama 3.1 8B Instruct
Meta logo

Llama 3.1 8B Instruct

by Meta · Released Jul 2024

Open Source
30.7
avg score
Rank #243
Compare
Better than 22% of all models
Context
131K tokens (~66 books)
Input $/1M
$0.02
Output $/1M
$0.03
Type
text
License
Open Source
Benchmarks
19 tested
Data as of
About

Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This 8B instruct-tuned version is fast and efficient. It has demonstrated strong performance compared to...

Tested on 19 benchmarks · BenchGecko score 30.7. Top scores: Chatbot Arena Elo — Overall (1211.2%), GSM8K (82.4%), PIQA (62.4%).

Capabilities
coding
19.7
#178 globally
reasoning
8.5
#187 globally
math
30.6
#188 globally
knowledge
23.1
#255 globally
general
17.9
#171 globally
language
50.6
#125 globally
Benchmark Scores
Compare All
Tested on 19 benchmarks · Ranked across 7 categories
Score Distribution (all 312 models)
0255075100
▲ You are here
codingCompare coding →
Aider — Code Editing

Code editing benchmark from the Aider project. Measures ability to apply targeted code changes while maintaining correctness and style.

37.6·
WeirdML

Unusual and adversarial machine learning challenges. Tests robustness of reasoning about edge cases in ML systems.

1.7·
reasoningCompare reasoning →
MUSR

HuggingFace MuSR (Multi-Step Reasoning). Tests multi-hop reasoning requiring chaining multiple facts together.

8.5·
mathCompare math →
GSM8K

Grade school math word problems. 8,500 problems testing multi-step arithmetic reasoning. A foundational math benchmark.

82.4·
MATH level 5

Competition-level math from AMC, AIME, and olympiad problems. Level 5 is the hardest tier, requiring creative problem-solving.

22.9·
MATH Level 5

HuggingFace evaluation of MATH Level 5 problems. Competition math requiring advanced reasoning and proof construction.

15.5·
Quick compare:Meta logovs Meta Llama 3 8B InstructMeta logovs Llama 3 70B Instructvs StarCoder 2 15B
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Model Family · Meta Llama 3.1
Jul 2024
43.8
n/an/actx22 benchmarks
Jul 2024
49.7+5.9
$0.40/M in131Kctx18 benchmarks
Jul 2024
30.7-19.0
$0.02/M in(-0.38)131Kctx19 benchmarks
See the full Llama 3.1 family →
Similar Models
Meta logo
Meta Llama 3 8B Instruct
Meta
30.7n/a
Meta logo
Llama 3 70B Instruct
Meta
30.6$0.51/1M
StarCoder 2 15B
Unknown
30.5n/a
Links
Info
MetaPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@Meta
Source Code
GitHubHugging Face
BenchGecko API
llama-3-1-8b-instruct
Specifications
  • Typetext
  • Context131K tokens (~66 books)
  • ReleasedJul 2024
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.000
Available On
Meta logoMeta$0.02
Categories
codingreasoningmathknowledgegenerallanguage
Learn More
context-windowtransformeropen weightstokens
Share & Export
Tweet
Related Models
Meta logoMeta Llama 3 8B Instruct
30.7
Meta logoLlama 3 70B Instruct
30.6
StarCoder 2 15B
30.5
Anthropic logoClaude 3 Sonnet
30.3
Llama 3.1 8B Instruct is an open-source text AI model by Meta, released in July 2024. It has an average benchmark score of 30.7. Context window: 131K tokens.

Related Models

Meta Llama 3 8B Instruct · MetaLlama 3 70B Instruct · MetaStarCoder 2 15B · UnknownClaude 3 Sonnet · Anthropicvicuna-13b-v1.1 · Unknown

Benchmarks

Chatbot Arena Elo — OverallGSM8KPIQAIFEvalMMLU

Related Pages

Meta · ProviderMeta · EconomyAll ModelsCompare ModelsPricingDevelopers · APIContext Window · Glossary

Key facts · as of 2026-10-05

  • Llama 3.1 8B Instruct by Meta. BenchGecko score 30.7, rank 243 of 312 scored models (normalized average of public benchmark scores).
  • List price $0.0200 input · $0.0300 output per 1M tokens (as of 2026-10-05).
  • Sold by 5 providers (as of 2026-10-05): DeepInfra (fp8) $0.0200 in / $0.0400 out · Novita (fp8) $0.0200 in / $0.0500 out · Groq $0.0500 in / $0.0800 out · Cloudflare (fp8) $0.15 in / $0.29 out · CoreWeave (bf16) $0.22 in / $0.22 out. Every provider

How to cite · data as of 2026-10-05

Llama 3.1 8B Instruct · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/model/llama-3-1-8b-instruct

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP