---
title: "Compare alternatives to DeepSeek V3 · Undominated.ai"
canonical: https://undominated.ai/alternatives/deepseek__deepseek-chat/
description: "48 alternatives score at least as high as DeepSeek V3 and cost no more under at least one listed workload. 41 preserve recorded capabilities; 7 have named trade-offs."
---

# Compare alternatives to DeepSeek V3 · Undominated.ai

> 48 alternatives score at least as high as DeepSeek V3 and cost no more under at least one listed workload. 41 preserve recorded capabilities; 7 have named trade-offs.

# Alternatives to DeepSeek V3

DeepSeek · LMArena 1332.5 · $0.500/M on a balanced workload · prices as of 2026-10-06

 [Compare models](https://undominated.ai/compare/) Comparison context

Choose up to three standard models in Compare.

Requirements and prompt length apply in Compare. This page keeps its stated price basis and workload controls.

## 41 models are higher-scoring or cheaper, with neither dimension worse.

Of the 144 models carrying both an independent score and a published price at the same delivery mode, **48** score at least as high as DeepSeek V3 and cost no more under at least one listed workload, with at least one of those dimensions improved. **41** match or beat its recorded context, maximum output, input modes, tool use and extended reasoning. **7** have a recorded capability loss, named on every row.

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 only while a candidate improves score or price without worsening the other under a listed workload. When that comparison no longer holds, it is not generated.

## Where it holds

| Workload | DeepSeek V3 $/M | Recorded capabilities preserved | Recorded capability losses |
| --- | --- | --- | --- |
| Balanced | $0.500 | 27 | 7 |
| Summarise | $0.329 | 36 | 7 |
| Chat | $0.629 | 34 | 7 |
| Code gen | $0.800 | 28 | 7 |
| Agentic | $0.415 | 41 | 7 |

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 | 1332.5 |
| --- | --- |
| Context window | 163K |
| Max output | 16K |
| Input modes | text |
| Tool use | yes |
| Extended reasoning | no |

Preserving recorded capabilities requires matching or beating these fields. [Full record for DeepSeek V3](/models/deepseek__deepseek-chat/), or [check DeepSeek V3 against the whole catalogue](/check/deepseek__deepseek-chat/).

## Recorded capabilities preserved 41

Each option has a higher score or a lower price, with neither dimension worse under the listed workloads. It matches or beats DeepSeek V3 on recorded context, maximum output, input modes, tool use and extended reasoning.

| # | Model | Intelligence | $/M | Holds under |
| --- | --- | --- | --- | --- |
| 1 | [GLM 5.3 Flash](/models/z-ai__glm-5.3-flash/) Z.ai | 1469.6 +137.1 | $0.238 | Balanced −53% Summarise −59% Chat −60% Code gen −56% Agentic −68% |
| 2 | [DeepSeek V4.1 Flash](/models/deepseek__deepseek-v4.1-flash/) DeepSeek | 1462.5 +130 | $0.182 | Balanced −64% Summarise −80% Chat −61% Code gen −55% Agentic −75% |
| 3 | [MiMo-V2.6-Flash](/models/xiaomi__mimo-v2.6-flash/) Xiaomi | 1456.4 +123.9 | $0.175 | Balanced −65% Summarise −67% Chat −75% Code gen −73% Agentic −81% |
| 4 | [Gemma 4 31B](/models/google__gemma-4-31b-it/) Google | 1443.4 +110.9 | $0.205 | Balanced −59% Summarise −53% Chat −61% Code gen −63% Agentic −57% |
| 5 | [Hy3](/models/tencent__hy3/) Tencent | 1440.7 +108.2 | $0.231 | Balanced −54% Summarise −62% Chat −59% Code gen −55% Agentic −68% |
| 6 | [Gemma 4 26B A4B](/models/google__gemma-4-26b-a4b-it/) Google | 1433.9 +101.4 | $0.127 | Balanced −75% Summarise −79% Chat −76% Code gen −74% Agentic −81% |
| 7 | [DeepSeek V4 Flash 0423](/models/deepseek__deepseek-v4-flash/) DeepSeek | 1432.1 +99.6 | $0.113 | Balanced −78% Summarise −78% Chat −83% Code gen −83% Agentic −85% |
| 8 | [MiMo-V2.5](/models/xiaomi__mimo-v2.5/) Xiaomi | 1427.5 +95 | $0.175 | Balanced −65% Summarise −67% Chat −75% Code gen −73% Agentic −81% |
| 9 | [MiniMax M3](/models/minimax__minimax-m3/) MiniMax | 1432.1 +99.6 | $0.485 | Balanced −3% Summarise −22% Chat −14% Code gen −6% Agentic −35% |
| 10 | [GPT-5.6 Luna](/models/openai__gpt-5.6-luna/) OpenAI | 1431 +98.5 | $0.450 | Balanced −10% Summarise −40% Chat −13% Code gen −2% Agentic −41% |
| 11 | [DeepSeek V3.2](/models/deepseek__deepseek-v3.2/) DeepSeek | 1424.5 +92 | $0.290 | Balanced −42% Summarise −30% Chat −57% Code gen −60% Agentic −52% |
| 12 | [DeepSeek V3.2 Exp](/models/deepseek__deepseek-v3.2-exp/) DeepSeek | 1422.4 +89.9 | $0.305 | Balanced −39% Summarise −16% Chat −48% Code gen −56% Agentic −30% |
| 13 | [Qwen3 235B A22B Instruct 2507](/models/qwen__qwen3-235b-a22b-2507/) Qwen | 1419.3 +86.8 | $0.205 | Balanced −59% Summarise −66% Chat −56% Code gen −54% Agentic −62% |
| 14 | [Qwen3 VL 235B A22B Instruct](/models/qwen__qwen3-vl-235b-a22b-instruct/) Qwen | 1420.1 +87.6 | $0.370 | Balanced −26% Summarise −37% Chat −29% Code gen −25% Agentic −40% |
| 15 | [Qwen3 Next 80B A3B Instruct](/models/qwen__qwen3-next-80b-a3b-instruct/) Qwen | 1417.3 +84.8 | $0.268 | Balanced −46% Summarise −60% Chat −41% Code gen −37% Agentic −52% |
| 16 | [DeepSeek V3.1](/models/deepseek__deepseek-chat-v3.1/) DeepSeek | 1419.7 +87.2 | $0.453 | Balanced −10% Summarise −7% Chat −11% Code gen −12% Agentic −8% |
| 17 | [DeepSeek V3.1 Terminus](/models/deepseek__deepseek-v3.1-terminus/) DeepSeek | 1417.1 +84.6 | $0.453 | Balanced −10% Summarise −7% Chat −11% Code gen −12% Agentic −8% |
| 18 | [Step 3.5 Flash](/models/stepfun__step-3.5-flash/) StepFun | 1403 +70.5 | $0.150 | Balanced −70% Summarise −67% Chat −71% Code gen −73% Agentic −69% |
| 19 | [Qwen3.5-Flash](/models/qwen__qwen3.5-flash-02-23/) Qwen | 1397 +64.5 | $0.114 | Balanced −77% Summarise −77% Chat −77% Code gen −77% Agentic −77% |
| 20 | [GPT-6 Luna](/models/openai__gpt-6-luna/) OpenAI | 1391.5 +59 | $0.200 | Balanced −60% Summarise −71% Chat −63% Code gen −58% Agentic −74% |
| 21 | [Qwen3.5-35B-A3B](/models/qwen__qwen3.5-35b-a3b/) Qwen | 1394.4 +61.9 | $0.355 | Balanced −29% Summarise −52% Chat −27% Code gen −19% Agentic −48% |
| 22 | [Qwen3 30B A3B Instruct 2507](/models/qwen__qwen3-30b-a3b-instruct-2507/) Qwen | 1384.1 +51.6 | $0.143 | Balanced −72% Summarise −69% Chat −72% Code gen −73% Agentic −71% |
| 23 | [DeepSeek V3 0324](/models/deepseek__deepseek-chat-v3-0324/) DeepSeek | 1375 +42.5 | $0.438 | Balanced −13% Summarise −13% Chat −13% Code gen −13% Agentic −13% |
| 24 | [GPT-5.4 Nano](/models/openai__gpt-5.4-nano/) OpenAI | 1372.1 +39.6 | $0.463 | Balanced −8% Summarise −39% Chat −10% Agentic −40% |
| 25 | [Qwen3 Next 80B A3B Thinking](/models/qwen__qwen3-next-80b-a3b-thinking/) Qwen | 1368.3 +35.8 | $0.413 | Balanced −18% Summarise −38% Chat −9% Code gen −3% Agentic −26% |
| 26 | [GLM 4.7 Flash](/models/z-ai__glm-4.7-flash/) Z.ai | 1350.9 +18.4 | $0.145 | Balanced −71% Summarise −81% Chat −71% Code gen −68% Agentic −80% |
| 27 | [Trinity Large Thinking](/models/arcee-ai__trinity-large-thinking/) Arcee | 1340 tie | $0.388 | Balanced −23% Summarise −32% Chat −34% Code gen −29% Agentic −47% |
| 28 | [MiMo-V2.6-Pro](/models/xiaomi__mimo-v2.6-pro/) Xiaomi | 1491 +158.5 | $0.478 chat | Chat −24% Code gen −18% Agentic −42% |
| 29 | [MiMo-V2.5-Pro](/models/xiaomi__mimo-v2.5-pro/) Xiaomi | 1465 +132.5 | $0.480 chat | Chat −24% Code gen −17% Agentic −41% |
| 30 | [Qwen3.7 Plus](/models/qwen__qwen3.7-plus/) Qwen | 1454.6 +122.1 | $0.295 summarise | Summarise −10% Chat under 1% less Agentic −25% |
| 31 | [Qwen3.8 27B](/models/qwen__qwen3.8-27b/) Qwen | 1440.7 +108.2 | $0.183 summarise | Summarise −44% Agentic −11% |
| 32 | [GLM 4.7](/models/z-ai__glm-4.7/) Z.ai | 1435.5 +103 | $0.412 agentic | Agentic −1% |
| 33 | [Gemini 3.1 Flash Lite Preview](/models/google__gemini-3.1-flash-lite-preview/) Google | 1415.7 +83.2 | $0.248 summarise | Summarise −24% Agentic −27% |
| 34 | [Inkling Small](/models/thinkingmachines__inkling-small/) Thinking Machines | 1413.7 +81.2 | $0.354 agentic | Agentic −15% |
| 35 | [Qwen3.5-27B](/models/qwen__qwen3.5-27b/) Qwen | 1408.9 +76.4 | $0.263 summarise | Summarise −20% Agentic −4% |
| 36 | [MiniMax M2.7](/models/minimax__minimax-m2.7/) MiniMax | 1404.6 +72.1 | $0.277 summarise | Summarise −16% Chat −6% Agentic −30% |
| 37 | [Solar Pro 4](/models/upstage__solar-pro4/) Upstage | 1386.2 +53.7 | $0.277 summarise | Summarise −16% Chat −6% Agentic −30% |
| 38 | [GPT-5 Mini](/models/openai__gpt-5-mini/) OpenAI | 1373.2 +40.7 | $0.273 summarise | Summarise −17% Agentic −9% |
| 39 | [MiniMax M2.5](/models/minimax__minimax-m2.5/) MiniMax | 1359.3 +26.8 | $0.273 summarise | Summarise −17% Chat −7% Agentic −30% |
| 40 | [MiniMax M2](/models/minimax__minimax-m2/) MiniMax | 1339.9 tie | $0.268 summarise | Summarise −18% Chat −8% Agentic −34% |
| 41 | [GPT-4.1 Mini](/models/openai__gpt-4.1-mini/) OpenAI | 1340.4 tie | $0.402 agentic | Agentic −3% |

## Alternatives with recorded capability losses 7

These improve score or price without worsening the other under the listed workloads, but reduce at least one recorded capability. Read the last column before switching.

| Model | Intelligence | $/M | Holds under | Recorded capability loss |
| --- | --- | --- | --- | --- |
| [GLM 4.5 Air](/models/z-ai__glm-4.5-air/) Z.ai | 1383.9 +51.4 | $0.310 | Balanced −38% Summarise −59% Chat −39% Code gen −31% Agentic −58% | 164K → 131K context |
| [GLM 4.6V](/models/z-ai__glm-4.6v/) Z.ai | 1376.8 +44.3 | $0.450 | Balanced −10% Summarise −21% Chat −26% Code gen −20% Agentic −42% | 164K → 131K context |
| [gpt-oss-120b](/models/openai__gpt-oss-120b/) OpenAI | 1365.4 +32.9 | $0.065 | Balanced −87% Summarise −89% Chat −86% Code gen −86% Agentic −88% | 164K → 131K context |
| [Gemma 3 27B](/models/google__gemma-3-27b-it/) Google | 1357.8 +25.3 | $0.100 | Balanced −80% Summarise −74% Chat −82% Code gen −84% Agentic −78% | 164K → 131K context |
| [Mercury 2](/models/inception__mercury-2/) Inception | 1355.2 +22.7 | $0.375 | Balanced −25% Summarise −36% Chat −39% Code gen −34% Agentic −54% | 164K → 128K context |
| [Qwen3 32B](/models/qwen__qwen3-32b/) Qwen | 1340.1 tie | $0.130 | Balanced −74% Summarise −73% Chat −75% Code gen −75% Agentic −73% | 164K → 131K context |
| [Gemma 3 12B](/models/google__gemma-3-12b-it/) Google | 1334.2 tie | $0.075 | Balanced −85% Summarise −83% Chat −86% Code gen −86% Agentic −84% | 164K → 131K 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 ±10.47 points, so a gap smaller than that is marked *tie* rather than an improvement — see [significance bands](/significance/).
 - 204 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.
 - Models with no higher-scoring or cheaper alternative that is no worse on the other axis are on the [value frontier](/frontier/). Models with such an alternative are [listed here](/alternatives/).

## Continue your investigation

 - [Check requirements](/check/)
- [Compare exact differences](/compare/)
- [Set a quality floor](/cheapest-at/)
- [Inspect tier boundaries](/cliffs/)