MMODELYST
Models/Llama 3.1 Instruct 8B
MDL

Llama 3.1 Instruct 8B

Meta

Verified against Artificial Analysis · Jul 13, 2026
Capability
31
percentile index
Price /M
$0.075 / $0.095
in / out
Context
Avg score
20.1
11 benchmarks
AA Intelligence
7.6
index
Speed
171 tok/s
TTFT 0.5s
Released
Jul 23, 2024
About

Llama 3.1 Instruct 8B is a open-weights model in the Llama family from Meta. Benchmarked on 11 evals, averaging 20.1.

Benchmark scores · 11
avg 20.1 — higher is better
MATH
51.9
MMLU-Pro
47.6
IFBench
28.6
GPQA Diamond
25.9
τ²-bench
16.4
AA Long-Context Reasoning
15.7
SciCode
13.2
LiveCodeBench
11.6
Humanity's Last Exam
5.1
AIME
4.3
Terminal-Bench Hard
0.8
Details
PublisherMeta
FamilyLlama
Parameters
License
WeightsOpen
ReleasedJul 23, 2024
Performance & price
Output speed171 tok/s
Latency (TTFT)0.5s
Price in / out$0.075 / $0.095 per 1M
Blended price$0.08 / 1M
Coding index5.4
Math index4.3

Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →

Capability over time

Append-only ledger — every observed change to this model's numbers.

API providers
Artificial Analysis$0.095/M