---
title: "Cheaper alternatives to Command R (08-2024) · Undominated.ai"
canonical: https://undominated.ai/alternatives/cohere__command-r-08-2024/
description: "32 models score at least as high as Command R (08-2024) and cost no more on a balanced workload. 27 give up nothing measurable; the other 5 name what they drop. Measured prices and independent scores."
---

# Cheaper alternatives to Command R (08-2024) · Undominated.ai

> 32 models score at least as high as Command R (08-2024) and cost no more on a balanced workload. 27 give up nothing measurable; the other 5 name what they drop. Measured prices and independent scores.

# Cheaper alternatives to Command R (08-2024)

Cohere · LMArena 1187.5 · $0.263/M on a balanced workload · prices as of 2026-08-28

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

Of the 132 models carrying both an independent score and a published price at the same delivery mode, **32** score at least as high as Command R (08-2024) and cost no more under at least one workload. **27** match or beat it on every capability we hold — context, maximum output, input modes, tool use and extended reasoning. **5** do 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 Command R (08-2024), it is not generated.

## Where it holds

| Workload | Command R (08-2024) $/M | Better and cheaper | With a trade |
| --- | --- | --- | --- |
| Balanced | $0.263 | 21 | 4 |
| Summarise | $0.172 | 23 | 4 |
| Chat | $0.330 | 23 | 4 |
| Code gen | $0.420 | 24 | 5 |
| Agentic | $0.217 | 24 | 4 |

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 | 1187.5 |
| --- | --- |
| Context window | 128K |
| Max output | 4K |
| Input modes | text |
| Tool use | yes |
| Extended reasoning | no |

A replacement has to clear every line above, not just the score. [Full record for Command R (08-2024)](/models/cohere__command-r-08-2024/).

## Better and cheaper, nothing given up 27

Each of these matches or beats Command R (08-2024) 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 | Gemma 4 31B Google | 1441.7 +254.2 | $0.152 | Balanced −42% Summarise −47% Chat −46% Code gen −44% Agentic −52% |
| 2 | Hy3 Tencent | 1441.2 +253.7 | $0.144 | Balanced −45% Summarise −55% Chat −51% Code gen −46% Agentic −62% |
| 3 | Gemma 4 26B A4B Google | 1434.6 +247.1 | $0.138 | Balanced −48% Summarise −52% Chat −46% Code gen −45% Agentic −49% |
| 4 | DeepSeek V4 Flash 0423 DeepSeek | 1431.6 +244.1 | $0.101 | Balanced −61% Summarise −61% Chat −72% Code gen −70% Agentic −75% |
| 5 | MiMo-V2.5 Xiaomi | 1427.3 +239.8 | $0.175 | Balanced −33% Summarise −37% Chat −53% Code gen −49% Agentic −63% |
| 6 | Qwen3 235B A22B Instruct 2507 Qwen | 1419.3 +231.8 | $0.205 | Balanced −22% Summarise −34% Chat −17% Code gen −13% Agentic −27% |
| 7 | Step 3.5 Flash StepFun | 1403.8 +216.3 | $0.150 | Balanced −43% Summarise −36% Chat −45% Code gen −48% Agentic −40% |
| 8 | Qwen3.5-Flash Qwen | 1397.6 +210.1 | $0.114 | Balanced −57% Summarise −57% Chat −57% Code gen −57% Agentic −57% |
| 9 | Qwen3 30B A3B Instruct 2507 Qwen | 1384.3 +196.8 | $0.084 | Balanced −68% Summarise −68% Chat −68% Code gen −68% Agentic −68% |
| 10 | Solar Pro 4 Upstage | 1376.2 +188.7 | $0.052 | Balanced −80% Summarise −84% Chat −82% Code gen −80% Agentic −87% |
| 11 | gpt-oss-120b OpenAI | 1365.6 +178.1 | $0.070 | Balanced −73% Summarise −75% Chat −73% Code gen −72% Agentic −74% |
| 12 | Gemma 3 27B Google | 1358.3 +170.8 | $0.172 | Balanced −34% Summarise −50% Chat −35% Code gen −29% Agentic −49% |
| 13 | GLM 4.7 Flash Z.ai | 1352.9 +165.4 | $0.145 | Balanced −45% Summarise −64% Chat −45% Code gen −38% Agentic −63% |
| 14 | Qwen3 32B Qwen | 1340.1 +152.6 | $0.130 | Balanced −50% Summarise −48% Chat −52% Code gen −52% Agentic −49% |
| 15 | Gemma 3 12B Google | 1334.2 +146.7 | $0.075 | Balanced −71% Summarise −68% Chat −73% Code gen −74% Agentic −70% |
| 16 | GPT-5 Nano OpenAI | 1320.3 +132.8 | $0.138 | Balanced −48% Summarise −68% Chat −47% Code gen −39% Agentic −65% |
| 17 | Qwen3 30B A3B Qwen | 1316.9 +129.4 | $0.215 | Balanced −18% Summarise −19% Chat −18% Code gen −17% Agentic −19% |
| 18 | Granite 4.1 8B IBM | 1291.6 +104.1 | $0.063 | Balanced −76% Summarise −70% Chat −79% Code gen −81% Agentic −74% |
| 19 | gpt-oss-20b OpenAI | 1287.8 +100.3 | $0.055 | Balanced −79% Summarise −80% Chat −79% Code gen −79% Agentic −79% |
| 20 | GPT-4.1 Nano OpenAI | 1284.8 +97.3 | $0.175 | Balanced −33% Summarise −46% Chat −40% Code gen −35% Agentic −54% |
| 21 | GPT-4o-mini (2024-07-18) OpenAI | 1286.6 +99.1 | $0.263 | Balanced same price Summarise −12% Chat −7% Code gen −1% Agentic −21% |
| 22 | DeepSeek V3.2 DeepSeek | 1424.6 +237.1 | $0.269 chat | Chat −18% Code gen −23% Agentic −8% |
| 23 | DeepSeek V3.2 Exp DeepSeek | 1424.4 +236.9 | $0.326 chat | Chat −1% Code gen −16% |
| 24 | Qwen3 Next 80B A3B Instruct Qwen | 1418.6 +231.1 | $0.141 summarise | Summarise −18% |
| 25 | GLM 4.5 Air Z.ai | 1382.8 +195.3 | $0.136 summarise | Summarise −21% Agentic −19% |
| 26 | Mercury 2 Inception | 1357.8 +170.3 | $0.191 agentic | Agentic −12% |
| 27 | Llama 3.1 70B Instruct Meta | 1261 +73.5 | $0.400 code gen | Code gen −5% |

## Cheaper and higher-scoring, but you give something up 5

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 |
| --- | --- | --- | --- | --- |
| Olmo 3 32B Think Allen AI | 1298.5 +111 | $0.237 | Balanced −10% Summarise −3% Chat −12% Code gen −14% Agentic −7% | 128K → 66K context no tool use |
| Gemma 3 4B Google | 1290.8 +103.3 | $0.063 | Balanced −76% Summarise −70% Chat −79% Code gen −81% Agentic −74% | no tool use |
| Mistral Small 3 Mistral | 1233.6 +46.1 | $0.058 | Balanced −78% Summarise −70% Chat −81% Code gen −84% Agentic −75% | 128K → 33K context no tool use |
| Phi 4 Microsoft | 1216.8 +29.3 | $0.087 | Balanced −67% Summarise −57% Chat −70% Code gen −73% Agentic −63% | 128K → 16K context no tool use |
| Qwen2.5 72B Instruct Qwen | 1269.1 +81.6 | $0.384 code gen | Code gen −9% | 128K → 33K context |

## 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/).