---
title: "Cheaper alternatives to DeepSeek V3 0324 · Undominated.ai"
canonical: https://undominated.ai/alternatives/deepseek__deepseek-chat-v3-0324/
description: "24 models score at least as high as DeepSeek V3 0324 and cost no more on a balanced workload. 5 give up nothing measurable; the other 19 name what they drop. Measured prices and independent scores."
---

# Cheaper alternatives to DeepSeek V3 0324 · Undominated.ai

> 24 models score at least as high as DeepSeek V3 0324 and cost no more on a balanced workload. 5 give up nothing measurable; the other 19 name what they drop. Measured prices and independent scores.

# Cheaper alternatives to DeepSeek V3 0324

DeepSeek · LMArena 1375 · $0.438/M on a balanced workload · prices as of 2026-08-28

## 5 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, **24** score at least as high as DeepSeek V3 0324 and cost no more under at least one workload. **5** match or beat it on every capability we hold — context, maximum output, input modes, tool use and extended reasoning. **19** 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 DeepSeek V3 0324, it is not generated.

## Where it holds

| Workload | DeepSeek V3 0324 $/M | Better and cheaper | With a trade |
| --- | --- | --- | --- |
| Balanced | $0.438 | 3 | 11 |
| Summarise | $0.287 | 4 | 17 |
| Chat | $0.550 | 3 | 14 |
| Code gen | $0.700 | 3 | 13 |
| Agentic | $0.362 | 5 | 17 |

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 | 1375 |
| --- | --- |
| Context window | 164K |
| Max output | 147K |
| Input modes | text |
| Tool use | yes |
| Extended reasoning | no |

A replacement has to clear every line above, not just the score. [Full record for DeepSeek V3 0324](/models/deepseek__deepseek-chat-v3-0324/).

## Better and cheaper, nothing given up 5

Each of these matches or beats DeepSeek V3 0324 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 | DeepSeek V4 Flash 0423 DeepSeek | 1431.6 +56.6 | $0.101 | Balanced −77% Summarise −77% Chat −83% Code gen −82% Agentic −85% |
| 2 | DeepSeek V3.2 DeepSeek | 1424.6 +49.6 | $0.290 | Balanced −34% Summarise −20% Chat −51% Code gen −54% Agentic −45% |
| 3 | Qwen3 Next 80B A3B Instruct Qwen | 1418.6 +43.6 | $0.350 | Balanced −20% Summarise −51% Chat −11% Code gen under 1% less Agentic −36% |
| 4 | MiniMax M3 MiniMax | 1434.8 +59.8 | $0.277 summarise | Summarise −4% Agentic −19% |
| 5 | Inkling Small Thinking Machines | 1411.7 +36.7 | $0.354 agentic | Agentic −2% |

## Cheaper and higher-scoring, but you give something up 19

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 |
| --- | --- | --- | --- | --- |
| Gemma 4 31B Google | 1441.7 +66.7 | $0.152 | Balanced −65% Summarise −68% Chat −68% Code gen −66% Agentic −71% | 147K → 16K max output |
| Hy3 Tencent | 1441.2 +66.2 | $0.144 | Balanced −67% Summarise −73% Chat −70% Code gen −68% Agentic −77% | 147K → 128K max output |
| Gemma 4 26B A4B Google | 1434.6 +59.6 | $0.138 | Balanced −69% Summarise −71% Chat −68% Code gen −67% Agentic −70% | 147K → 16K max output |
| MiMo-V2.5 Xiaomi | 1427.3 +52.3 | $0.175 | Balanced −60% Summarise −62% Chat −72% Code gen −70% Agentic −78% | 147K → 131K max output |
| DeepSeek V3.2 Exp DeepSeek | 1424.4 +49.4 | $0.305 | Balanced −30% Summarise −4% Chat −41% Code gen −49% Agentic −20% | 147K → 66K max output |
| Qwen3 235B A22B Instruct 2507 Qwen | 1419.3 +44.3 | $0.205 | Balanced −53% Summarise −61% Chat −50% Code gen −48% Agentic −56% | 147K → 16K max output |
| Step 3.5 Flash StepFun | 1403.8 +28.8 | $0.150 | Balanced −66% Summarise −62% Chat −67% Code gen −69% Agentic −64% | 147K → 66K max output |
| Qwen3.5-Flash Qwen | 1397.6 +22.6 | $0.114 | Balanced −74% Summarise −74% Chat −74% Code gen −74% Agentic −74% | 147K → 66K max output |
| Qwen3 30B A3B Instruct 2507 Qwen | 1384.3 tie | $0.084 | Balanced −81% Summarise −81% Chat −81% Code gen −81% Agentic −81% | 147K → 32K max output |
| GLM 4.5 Air Z.ai | 1382.8 tie | $0.310 | Balanced −29% Summarise −53% Chat −30% Code gen −21% Agentic −52% | 164K → 131K context 147K → 98K max output |
| Solar Pro 4 Upstage | 1376.2 tie | $0.052 | Balanced −88% Summarise −90% Chat −89% Code gen −88% Agentic −92% | 147K → 131K max output |
| MiMo-V2.5-Pro Xiaomi | 1465 +90 | $0.480 chat | Chat −13% Code gen −5% Agentic −33% | 147K → 131K max output |
| Qwen3.7 Plus Qwen | 1456.2 +81.2 | $0.312 agentic | Agentic −14% | 147K → 131K max output |
| GPT-5.6 Luna OpenAI | 1428.5 +53.5 | $0.199 summarise | Summarise −31% Chat −1% Agentic −33% | 147K → 128K max output |
| Qwen3 VL 235B A22B Instruct Qwen | 1420.9 +45.9 | $0.263 summarise | Summarise −8% | 147K → 33K max output |
| DeepSeek V3.1 Terminus DeepSeek | 1419.6 +44.6 | $0.268 summarise | Summarise −7% Chat −5% Code gen under 1% less Agentic −17% | 147K → 33K max output |
| Gemini 3.1 Flash Lite Preview Google | 1414.8 +39.8 | $0.248 summarise | Summarise −14% Agentic −16% | 147K → 66K max output |
| Qwen3.5-27B Qwen | 1407.9 +32.9 | $0.263 summarise | Summarise −8% | 147K → 66K max output |
| MiniMax M2.7 MiniMax | 1405.3 +30.3 | $0.277 summarise | Summarise −4% Agentic −19% | 147K → 131K max output |

## 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/).