---
title: "Compare alternatives to Mercury 2 · Undominated.ai"
canonical: https://undominated.ai/alternatives/inception__mercury-2/
description: "25 alternatives score at least as high as Mercury 2 and cost no more under at least one listed workload. 17 preserve recorded capabilities; 8 have named trade-offs."
---

# Compare alternatives to Mercury 2 · Undominated.ai

> 25 alternatives score at least as high as Mercury 2 and cost no more under at least one listed workload. 17 preserve recorded capabilities; 8 have named trade-offs.

# Alternatives to Mercury 2

Inception · LMArena 1355.2 · $0.375/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.

## 17 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, **25** score at least as high as Mercury 2 and cost no more under at least one listed workload, with at least one of those dimensions improved. **17** match or beat its recorded context, maximum output, input modes, tool use and extended reasoning. **8** 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 | Mercury 2 $/M | Recorded capabilities preserved | Recorded capability losses |
| --- | --- | --- | --- |
| Balanced | $0.375 | 14 | 7 |
| Summarise | $0.211 | 15 | 8 |
| Chat | $0.383 | 13 | 5 |
| Code gen | $0.532 | 13 | 5 |
| Agentic | $0.191 | 12 | 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 | 1355.2 |
| --- | --- |
| Context window | 128K |
| Max output | 50K |
| Input modes | text |
| Tool use | yes |
| Extended reasoning | yes |

Preserving recorded capabilities requires matching or beating these fields. [Full record for Mercury 2](/models/inception__mercury-2/), or [check Mercury 2 against the whole catalogue](/check/inception__mercury-2/).

## Recorded capabilities preserved 17

Each option has a higher score or a lower price, with neither dimension worse under the listed workloads. It matches or beats Mercury 2 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 +114.4 | $0.238 | Balanced −37% Summarise −37% Chat −34% Code gen −34% Agentic −31% |
| 2 | [DeepSeek V4.1 Flash](/models/deepseek__deepseek-v4.1-flash/) DeepSeek | 1462.5 +107.3 | $0.182 | Balanced −51% Summarise −68% Chat −35% Code gen −32% Agentic −45% |
| 3 | [MiMo-V2.6-Flash](/models/xiaomi__mimo-v2.6-flash/) Xiaomi | 1456.4 +101.2 | $0.175 | Balanced −53% Summarise −49% Chat −60% Code gen −60% Agentic −58% |
| 4 | [Hy3](/models/tencent__hy3/) Tencent | 1440.7 +85.5 | $0.231 | Balanced −38% Summarise −41% Chat −32% Code gen −32% Agentic −31% |
| 5 | [Gemma 4 26B A4B](/models/google__gemma-4-26b-a4b-it/) Google | 1433.9 +78.7 | $0.127 | Balanced −66% Summarise −67% Chat −61% Code gen −62% Agentic −58% |
| 6 | [DeepSeek V4 Flash 0423](/models/deepseek__deepseek-v4-flash/) DeepSeek | 1432.1 +76.9 | $0.113 | Balanced −70% Summarise −65% Chat −73% Code gen −74% Agentic −68% |
| 7 | [MiMo-V2.5](/models/xiaomi__mimo-v2.5/) Xiaomi | 1427.5 +72.3 | $0.175 | Balanced −53% Summarise −49% Chat −60% Code gen −60% Agentic −58% |
| 8 | [DeepSeek V3.2](/models/deepseek__deepseek-v3.2/) DeepSeek | 1424.5 +69.3 | $0.290 | Balanced −23% Chat −30% Code gen −40% |
| 9 | [DeepSeek V3.2 Exp](/models/deepseek__deepseek-v3.2-exp/) DeepSeek | 1422.4 +67.2 | $0.305 | Balanced −19% Chat −15% Code gen −33% |
| 10 | [Step 3.5 Flash](/models/stepfun__step-3.5-flash/) StepFun | 1403 +47.8 | $0.150 | Balanced −60% Summarise −48% Chat −53% Code gen −59% Agentic −32% |
| 11 | [Qwen3.5-Flash](/models/qwen__qwen3.5-flash-02-23/) Qwen | 1397 +41.8 | $0.114 | Balanced −70% Summarise −65% Chat −63% Code gen −66% Agentic −51% |
| 12 | [GPT-6 Luna](/models/openai__gpt-6-luna/) OpenAI | 1391.5 +36.3 | $0.200 | Balanced −47% Summarise −55% Chat −39% Code gen −37% Agentic −44% |
| 13 | [GLM 4.5 Air](/models/z-ai__glm-4.5-air/) Z.ai | 1383.9 +28.7 | $0.310 | Balanced −17% Summarise −35% Agentic −8% |
| 14 | [gpt-oss-120b](/models/openai__gpt-oss-120b/) OpenAI | 1365.4 tie | $0.065 | Balanced −83% Summarise −82% Chat −78% Code gen −79% Agentic −73% |
| 15 | [Qwen3.8 27B](/models/qwen__qwen3.8-27b/) Qwen | 1440.7 +85.5 | $0.183 summarise | Summarise −13% |
| 16 | [GPT-5.6 Luna](/models/openai__gpt-5.6-luna/) OpenAI | 1431 +75.8 | $0.199 summarise | Summarise −6% |
| 17 | [GPT-5.4 Nano](/models/openai__gpt-5.4-nano/) OpenAI | 1372.1 +16.9 | $0.201 summarise | Summarise −5% |

## Alternatives with recorded capability losses 8

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 |
| --- | --- | --- | --- | --- |
| [Gemma 4 31B](/models/google__gemma-4-31b-it/) Google | 1443.4 +88.2 | $0.205 | Balanced −45% Summarise −27% Chat −36% Code gen −44% Agentic −6% | 50K → 16K max output |
| [Qwen3 235B A22B Instruct 2507](/models/qwen__qwen3-235b-a22b-2507/) Qwen | 1419.3 +64.1 | $0.205 | Balanced −45% Summarise −46% Chat −28% Code gen −31% Agentic −17% | 50K → 16K max output no extended reasoning |
| [Qwen3 VL 235B A22B Instruct](/models/qwen__qwen3-vl-235b-a22b-instruct/) Qwen | 1420.1 +64.9 | $0.370 | Balanced −1% Summarise −1% | 50K → 33K max output no extended reasoning |
| [Qwen3 Next 80B A3B Instruct](/models/qwen__qwen3-next-80b-a3b-instruct/) Qwen | 1417.3 +62.1 | $0.268 | Balanced −29% Summarise −38% Chat −3% Code gen −5% | no extended reasoning |
| [Qwen3.5-35B-A3B](/models/qwen__qwen3.5-35b-a3b/) Qwen | 1394.4 +39.2 | $0.355 | Balanced −5% Summarise −25% | 50K → 33K max output |
| [Qwen3 30B A3B Instruct 2507](/models/qwen__qwen3-30b-a3b-instruct-2507/) Qwen | 1384.1 +28.9 | $0.143 | Balanced −62% Summarise −52% Chat −55% Code gen −59% Agentic −36% | no extended reasoning |
| [Gemma 3 27B](/models/google__gemma-3-27b-it/) Google | 1357.8 tie | $0.100 | Balanced −73% Summarise −60% Chat −71% Code gen −76% Agentic −52% | no extended reasoning |
| [Qwen3 Next 80B A3B Thinking](/models/qwen__qwen3-next-80b-a3b-thinking/) Qwen | 1368.3 +13.1 | $0.203 summarise | Summarise −4% | 50K → 33K 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 ±10.47 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/)