Lepton AI
Efficient

Mixtral 8x7B (Lepton)

MoE

The original Mixtral MoE from late 2023. Kicked off the open-weights MoE wave; now superseded by 8x22B and Llama 3.x.

Mixtral 8x7B (Lepton) is a efficient AI model from Lepton AI. It costs $0.5000 per million input tokens and $0.5000 per million output tokens (blended $0.5000/M), with a 32,000-token context window.

Profile inherited from upstream Mixtral 8x7B ↗ — this is a hosted variant of the same open-weights model.

RatesUSD per million tokens. Blend is 70% input, 30% output, our calculation.

INPUT
$0.5000/M
per million input tokens
OUTPUT
$0.5000/M
per million output tokens
BLENDED 70/30
$0.5000/M
unchanged since 3 May
CONTEXT
32,000
tokens

Hand-maintained from Lepton AI's published pricing; no check date on record.

What it is good at
  • Open weights
  • 32K context
  • Wide ecosystem
Typical use cases
  • Legacy Mixtral 8x7B deployments
  • Fine-tune base

Benchmarks

vs. best public score
Scores inherited from Mixtral 8x7B — this is a hosted variant of the same open-weights model, so the underlying benchmark scores are identical.
MMLU70.6%
Multitask academic knowledge across 57 subjects. Source: Mistral AI (self-reported), MMLU (Mixtral 8x7B base model), read 2026-11-10.
MATH28.4%
High-school competition math problems. Source: Mistral AI (self-reported), MATH, read 2026-11-10.
HumanEval40.2%
Python function synthesis from docstrings. Source: Mistral AI (self-reported), HumanEval (0-shot), base model, read 2026-11-10.
LMArena Elo1197 Elo
Crowd-sourced head-to-head preference Elo rating. Source: LMArena (independent), LMArena Text Arena score, read 2026-11-10.
Hand-curated from each provider's published reports and public leaderboards. Methodology varies across sources, treat as directional rather than authoritative.

More from Lepton AI

See all 3 →

Frequently asked questions

How much does Mixtral 8x7B (Lepton) cost?

Mixtral 8x7B (Lepton) costs $0.5000 per million input tokens and $0.5000 per million output tokens, for a blended reference rate of $0.5000 per million tokens.

What is Mixtral 8x7B (Lepton)'s context window?

Mixtral 8x7B (Lepton) supports up to 32,000 tokens of context in a single request.

What is Mixtral 8x7B (Lepton) best for?

Mixtral 8x7B (Lepton) is well suited to Open weights, 32K context and Wide ecosystem.

Who makes Mixtral 8x7B (Lepton)?

Mixtral 8x7B (Lepton) is developed and served by Lepton AI. It was released in Dec 2023.

Terms used on this page