Alternatives to Solar Pro 4

Upstage LMArena 1386.2 $0.525/M on a balanced workload prices as of 2026-10-06

9 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, 27 score at least as high as Solar Pro 4 and cost no more under at least one listed workload, with at least one of those dimensions improved. 26 of them cost less; the rest match the price and have a higher score. 9 match or beat its recorded context, maximum output, input modes, tool use and extended reasoning. 18 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	Solar Pro 4 $/M	Recorded capabilities preserved	Recorded capability losses
Balanced	$0.525	6	16
Summarise	$0.277	7	15
Chat	$0.588	8	16
Code gen	$0.821	8	16
Agentic	$0.292	8	14

Workload mixes are defined on the methodology page. 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	1386.2
Context window	524K
Max output	131K
Input modes	text
Tool use	yes
Extended reasoning	yes

Preserving recorded capabilities requires matching or beating these fields. Full record for Solar Pro 4, or check Solar Pro 4 against the whole catalogue.

Recorded capabilities preserved 9

Each option has a higher score or a lower price, with neither dimension worse under the listed workloads. It matches or beats Solar Pro 4 on recorded context, maximum output, input modes, tool use and extended reasoning.

#	Model	Intelligence	$/M	Holds under
1	GLM 5.3 Flash Z.ai	1469.6 +83.4	$0.238	Balanced −55% Summarise −52% Chat −57% Code gen −57% Agentic −55%
2	DeepSeek V4.1 Flash DeepSeek	1462.5 +76.3	$0.182	Balanced −65% Summarise −76% Chat −58% Code gen −56% Agentic −64%
3	MiMo-V2.6-Flash Xiaomi	1456.4 +70.2	$0.175	Balanced −67% Summarise −61% Chat −74% Code gen −74% Agentic −73%
4	DeepSeek V4 Flash 0423 DeepSeek	1432.1 +45.9	$0.113	Balanced −79% Summarise −73% Chat −82% Code gen −83% Agentic −79%
5	MiMo-V2.5 Xiaomi	1427.5 +41.3	$0.175	Balanced −67% Summarise −61% Chat −74% Code gen −74% Agentic −73%
6	MiniMax M3 MiniMax	1432.1 +45.9	$0.485	Balanced −8% Summarise −7% Chat −8% Code gen −8% Agentic −8%
7	MiMo-V2.6-Pro Xiaomi	1491 +104.8	$0.478 chat	Chat −19% Code gen −20% Agentic −17%
8	MiMo-V2.5-Pro Xiaomi	1465 +78.8	$0.480 chat	Chat −18% Code gen −19% Agentic −17%
9	Qwen3.8 27B Qwen	1440.7 +54.5	$0.183 summarise	Summarise −34%

Alternatives with recorded capability losses 18

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 Google	1443.4 +57.2	$0.205	Balanced −61% Summarise −45% Chat −59% Code gen −64% Agentic −39%	524K → 262K context 131K → 16K max output
Hy3 Tencent	1440.7 +54.5	$0.231	Balanced −56% Summarise −55% Chat −56% Code gen −56% Agentic −55%	524K → 262K context 131K → 128K max output
Gemma 4 26B A4B Google	1433.9 +47.7	$0.127	Balanced −76% Summarise −75% Chat −75% Code gen −75% Agentic −73%	524K → 262K context
GPT-5.6 Luna OpenAI	1431 +44.8	$0.450	Balanced −14% Summarise −28% Chat −7% Code gen −4% Agentic −17%	131K → 128K max output
DeepSeek V3.2 DeepSeek	1424.5 +38.3	$0.290	Balanced −45% Summarise −17% Chat −54% Code gen −61% Agentic −31%	524K → 164K context 131K → 66K max output
DeepSeek V3.2 Exp DeepSeek	1422.4 +36.2	$0.305	Balanced −42% Chat −45% Code gen −57% Agentic under 1% less	524K → 164K context
Qwen3 235B A22B Instruct 2507 Qwen	1419.3 +33.1	$0.205	Balanced −61% Summarise −59% Chat −53% Code gen −55% Agentic −46%	524K → 262K context 131K → 16K max output no extended reasoning
Qwen3 VL 235B A22B Instruct Qwen	1420.1 +33.9	$0.370	Balanced −30% Summarise −25% Chat −24% Code gen −27% Agentic −15%	524K → 262K context 131K → 33K max output no extended reasoning
Qwen3 Next 80B A3B Instruct Qwen	1417.3 +31.1	$0.268	Balanced −49% Summarise −52% Chat −37% Code gen −38% Agentic −32%	524K → 262K context no extended reasoning
DeepSeek V3.1 DeepSeek	1419.7 +33.5	$0.453	Balanced −14% Chat −4% Code gen −14%	524K → 164K context 131K → 33K max output
DeepSeek V3.1 Terminus DeepSeek	1417.1 +30.9	$0.453	Balanced −14% Chat −4% Code gen −14%	524K → 164K context
Step 3.5 Flash StepFun	1403 +16.8	$0.150	Balanced −71% Summarise −60% Chat −69% Code gen −73% Agentic −56%	524K → 262K context 131K → 66K max output
Qwen3.5-Flash Qwen	1397 +10.8	$0.114	Balanced −78% Summarise −73% Chat −76% Code gen −78% Agentic −68%	131K → 66K max output
MiniMax M2.7 MiniMax	1404.6 +18.4	$0.525	Balanced same price Summarise same price Chat same price Code gen same price Agentic same price	524K → 205K context
GPT-6 Luna OpenAI	1391.5 tie	$0.200	Balanced −62% Summarise −66% Chat −60% Code gen −59% Agentic −64%	131K → 128K max output
Qwen3.5-35B-A3B Qwen	1394.4 tie	$0.355	Balanced −32% Summarise −43% Chat −22% Code gen −21% Agentic −26%	524K → 262K context 131K → 33K max output
Gemini 3.1 Flash Lite Preview Google	1415.7 +29.5	$0.248 summarise	Summarise −10%	131K → 66K max output
Qwen3.5-27B Qwen	1408.9 +22.7	$0.263 summarise	Summarise −5%	524K → 262K context 131K → 66K 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.
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. Models with such an alternative are listed here.

Choose models

Compare costs

Build with AI

Understand the numbers

Check our work

Reuse the data