← All Tools

Best Claude Sonnet 4.5 Alternatives

Claude Sonnet 4.5 by Anthropic is a mid-tier model priced at $3/15 per 1M tokens (in/out). It's on the expensive side — there are cheaper options with similar quality.

Claude Sonnet 4.5

AnthropicMid-Tier

Input

$3/1M

Output

$15/1M

Context

200K

Max Output

16K

Why Switch from Claude Sonnet 4.5?

Superseded by Sonnet 4.6
Same price as the newer model

Top Alternatives

#1Gemini 2.5 ProGoogleMid-Tier

38% cheaper, higher benchmark scores, 1M context (5x more).

Input

$1.25/1M

58% cheaper

Output

$10/1M

33% cheaper

Context

1M

Max Output

66K

MMLU-Pro: 87.5%(+3.0%)HumanEval: 93.5%(+0.5%)GPQA: 76%(+7.8%)
#2Claude Sonnet 4.6AnthropicMid-Tier

Comparable performance.

Input

$3/1M

Same price

Output

$15/1M

Same price

Context

200K

Max Output

16K

MMLU-Pro: 86%(+1.5%)HumanEval: 94%(+1.0%)GPQA: 70%(+1.8%)
#3GPT-5.2 CodexOpenAIFlagship

13% cheaper, higher benchmark scores, 66K max output.

Input

$1.75/1M

42% cheaper

Output

$14/1M

7% cheaper

Context

200K

Max Output

66K

MMLU-Pro: 89%(+4.5%)HumanEval: 95.5%(+2.5%)GPQA: 76%(+7.8%)
#4Mistral Large 3MistralFlagship

Dramatically cheaper (61% less), higher benchmark scores.

Input

$2/1M

33% cheaper

Output

$5/1M

67% cheaper

Context

128K

Max Output

16K

MMLU-Pro: 83%(-1.5%)HumanEval: 91%(-2.0%)GPQA:
#5GLM-4.7Zhipu AIMid-Tier

Dramatically cheaper (84% less), higher benchmark scores, 128K max output.

Input

$0.6/1M

80% cheaper

Output

$2.2/1M

85% cheaper

Context

200K

Max Output

128K

MMLU-Pro: 84.3%(-0.2%)HumanEval: GPQA: 85.7%(+17.5%)
#6GPT-4oOpenAIMid-Tier

31% cheaper, adds audio.

Input

$2.5/1M

17% cheaper

Output

$10/1M

33% cheaper

Context

128K

Max Output

16K

MMLU-Pro: 80.5%(-4.0%)HumanEval: 91%(-2.0%)GPQA: 64.2%(-4.0%)
#7GPT-5OpenAIFlagship

38% cheaper, higher benchmark scores, adds audio.

Input

$1.25/1M

58% cheaper

Output

$10/1M

33% cheaper

Context

128K

Max Output

16K

MMLU-Pro: 88.5%(+4.0%)HumanEval: 95%(+2.0%)GPQA: 73.5%(+5.3%)
#8Gemini 3.1 ProGoogleFlagship

22% cheaper, higher benchmark scores, 1M context (5x more).

Input

$2/1M

33% cheaper

Output

$12/1M

20% cheaper

Context

1M

Max Output

64K

MMLU-Pro: 91%(+6.5%)HumanEval: 95%(+2.0%)GPQA: 94.3%(+26.1%)

Full Comparison Table

ModelInput $/1MOutput $/1MContextMMLU-ProHumanEvalScore
Gemini 2.5 ProGoogle$1.2558% cheaper$10.0033% cheaper1M87.5%+3.0%93.5%+0.5%95
Claude Sonnet 4.6Anthropic$3.00Same price$15.00Same price200K86%+1.5%94%+1.0%90
GPT-5.2 CodexOpenAI$1.7542% cheaper$14.007% cheaper200K89%+4.5%95.5%+2.5%85
Mistral Large 3Mistral$2.0033% cheaper$5.0067% cheaper128K83%-1.5%91%-2.0%85
GLM-4.7Zhipu AI$0.6080% cheaper$2.2085% cheaper200K84.3%-0.2%84
GPT-4oOpenAI$2.5017% cheaper$10.0033% cheaper128K80.5%-4.0%91%-2.0%83
GPT-5OpenAI$1.2558% cheaper$10.0033% cheaper128K88.5%+4.0%95%+2.0%80
Gemini 3.1 ProGoogle$2.0033% cheaper$12.0020% cheaper1M91%+6.5%95%+2.0%80
Gemini 3 ProGoogle$2.0033% cheaper$12.0020% cheaper1M89.8%+5.3%94%+1.0%80
Mistral Medium 3Mistral$0.4087% cheaper$2.0087% cheaper128K76%-8.5%87%-6.0%77
Claude Opus 4.6Anthropic$5.0067% more$25.0067% more200K89.5%+5.0%95%+2.0%75
GPT-5.3 CodexOpenAI$2.0033% cheaper$16.007% more200K90%+5.5%96.5%+3.5%75
o4-miniOpenAI$1.1063% cheaper$4.4071% cheaper200K85%+0.5%93.5%+0.5%75
o3OpenAI$0.4087% cheaper$1.6089% cheaper200K87%+2.5%94.5%+1.5%75
Gemini 3 FlashGoogle$0.5083% cheaper$3.0080% cheaper1M78%-6.5%90%-3.0%75
Grok 4xAI$3.00Same price$15.00Same price128K86%+1.5%93%Same70
GLM-5Zhipu AI$1.0067% cheaper$3.2079% cheaper200K70.4%-14.1%91%-2.0%68
Gemini 2.5 FlashGoogle$0.1595% cheaper$0.6096% cheaper1M76%-8.5%89.5%-3.5%65
DeepSeek R1DeepSeek$0.5582% cheaper$2.1985% cheaper128K84%-0.5%92%-1.0%63
Claude Haiku 4.5Anthropic$0.8073% cheaper$4.0073% cheaper200K69.4%-15.1%88.1%-4.9%62
Llama 4 MaverickMeta$0.3190% cheaper$0.8594% cheaper1M80.5%-4.0%90.2%-2.8%53
DeepSeek V3DeepSeek$0.1495% cheaper$0.2898% cheaper164K78%-6.5%89%-4.0%53
MiniMax M2.5MiniMax$0.3090% cheaper$1.2092% cheaper200K82%-2.5%90%-3.0%53
GPT-4o MiniOpenAI$0.1595% cheaper$0.6096% cheaper128K68%-16.5%87.2%-5.8%52
Llama 4 ScoutMeta$0.1894% cheaper$0.6396% cheaper10M74.2%-10.3%86%-7.0%46

Head-to-Head Comparisons

Alternatives for Other Models