---
title: "Cheaper alternatives to Llama 3.1 8B Instruct · Undominated.ai"
canonical: https://undominated.ai/alternatives/meta-llama__llama-3.1-8b-instruct/
description: "4 models score at least as high as Llama 3.1 8B Instruct and cost no more on a balanced workload. 3 give up nothing measurable; the other 1 name what they drop. Measured prices and independent scores."
---

# Cheaper alternatives to Llama 3.1 8B Instruct · Undominated.ai

> 4 models score at least as high as Llama 3.1 8B Instruct and cost no more on a balanced workload. 3 give up nothing measurable; the other 1 name what they drop. Measured prices and independent scores.

# Cheaper alternatives to Llama 3.1 8B Instruct

Meta · LMArena 1186.7 · $0.058/M on a balanced workload · prices as of 2026-08-28

## 3 models are both better and cheaper, giving up nothing.

Of the 133 models carrying both an independent score and a published price at the same delivery mode, **4** score at least as high as Llama 3.1 8B Instruct and cost no more under at least one workload. **3** of them are genuinely cheaper; the rest match the price and win on score alone. **3** match or beat it on every capability we hold — context, maximum output, input modes, tool use and extended reasoning. **1** does not, and every row names what it drops.

Dominance is not a property of a model. It is a property of a model, a capability lens and a workload, and all three are named on every row. Scores are LMArena Elo, used under CC BY 4.0 from the official dataset; prices are blended per million tokens. A higher score is not a drop-in replacement.

This page is built from that comparison and nothing else. When no model is both better and cheaper than Llama 3.1 8B Instruct, it is not generated.

## Where it holds

| Workload | Llama 3.1 8B Instruct $/M | Better and cheaper | With a trade |
| --- | --- | --- | --- |
| Balanced | $0.058 | 2 | 1 |
| Summarise | $0.044 | 3 | 0 |
| Chat | $0.054 | 0 | 0 |
| Code gen | $0.066 | 0 | 0 |
| Agentic | $0.040 | 1 | 0 |

Workload mixes are defined on the [methodology page](/methodology/). Cached input is priced at the cached rate, and reasoning tokens at the worse of the reasoning and output rates.

## What you would be replacing

| Intelligence | 1186.7 |
| --- | --- |
| Context window | 131K |
| Max output | 118K |
| Input modes | text |
| Tool use | yes |
| Extended reasoning | no |

A replacement has to clear every line above, not just the score. [Full record for Llama 3.1 8B Instruct](/models/meta-llama__llama-3.1-8b-instruct/).

## Better and cheaper, nothing given up 3

Each of these matches or beats Llama 3.1 8B Instruct on context, maximum output, input modes, tool use and reasoning, scores at least as high, and costs no more.

| # | Model | Intelligence | $/M | Holds under |
| --- | --- | --- | --- | --- |
| 1 | Solar Pro 4 Upstage | 1376.2 +189.5 | $0.052 | Balanced −9% Summarise −38% Agentic −26% |
| 2 | gpt-oss-20b OpenAI | 1287.8 +101.1 | $0.055 | Balanced −4% Summarise −21% |
| 3 | gpt-oss-120b OpenAI | 1365.6 +178.9 | $0.044 summarise | Summarise −2% |

## Cheaper and higher-scoring, but you give something up 1

These score at least as high and cost no more on the two plotted axes, and lose something that is not on them. Read the last column before switching.

| Model | Intelligence | $/M | Holds under | What you give up |
| --- | --- | --- | --- | --- |
| Mistral Small 3 Mistral | 1233.6 +46.9 | $0.058 | Balanced same price | 131K → 33K context 118K → 16K max output no tool use |

## What this compares, and what it leaves out

 - Quality is LMArena Elo, used under CC BY 4.0 from the official dataset. The 95% confidence interval on a difference between two scores is about ±10.68 points, so a gap smaller than that is marked *tie* rather than an improvement — see [significance bands](/significance/).
 - 188 further models at this delivery mode carry a price but no independent score. They are absent from the comparison in both directions — unrated is not a zero, and an unmeasured model is neither an alternative nor a worse buy.
 - Retired models are never offered as an alternative, and a model only competes against its own delivery mode: batch trades latency for price, so it is not a like-for-like swap.
 - Nothing here measures latency, throughput, rate limits or how a model behaves on your prompts. Two models with the same index score are not interchangeable.
 - The models that nothing beats on both axes are on the [value frontier](/frontier/), and every other model something cheaper beats is [listed here](/alternatives/).