TokenCostEstimate bill →
Budget finder

The cheapest LLM APIs, ranked for your workload.

“Cheapest” depends on whether you send more than you generate. Pick your workload mix and every model re-ranks by the blended price you’ll actually pay.

Workload

50% input · 50% output. Ranked by blended price per 1M tokens for that mix.

#ModelProviderInputOutputBlended /1M
1
Qwen3.7 FlashCHEAPESTBudget
Qwen$0.03$0.13$0.08
2
Qwen3.5-9BBudget
Qwen$0.10$0.15$0.13
3
GLM 5.3 Flash (batch)Budget
Z.AI$0.06$0.20$0.13
4
DeepSeek V4 Flash 0423Budget
DeepSeek$0.09$0.18$0.14
5
Claude Haiku 5.5 (batch)Budget
Anthropic$0.05$0.25$0.15
6
GPT-6 Luna (batch)Budget
OpenAI$0.05$0.25$0.15
7
GPT-6 Luna Pro (batch)Budget
OpenAI$0.05$0.25$0.15
8
Qwen3.5-FlashBudget
Qwen$0.07$0.26$0.16
9
Mistral Small 4 (batch)Budget
Mistral$0.07$0.30$0.19
10
Gemma 4 26B A4B Budget
Google$0.09$0.30$0.20
11
DeepSeek V4.1 Flash (batch)Budget
DeepSeek$0.11$0.34$0.22
12
GLM 4.7 FlashBudget
Z.AI$0.06$0.40$0.23
13
Claude Haiku 5.5Budget
Anthropic$0.10$0.50$0.30
14
GPT-6 LunaBudget
OpenAI$0.10$0.50$0.30
15
GPT-6 Luna ProBudget
OpenAI$0.10$0.50$0.30
16
Gemma 4 31BBudget
Google$0.20$0.40$0.30
17
Qwen3.8 FlashBudget
Qwen$0.15$0.47$0.31
18
Qwen3.8 Omni FlashBudget
Qwen$0.15$0.47$0.31
19
GLM 5.3 FlashBudget
Z.AI$0.15$0.50$0.33
20
DeepSeek V4 Flash 0731Budget
DeepSeek$0.02$0.64$0.33
21
GPT-5.6 Luna (batch)Budget
OpenAI$0.10$0.60$0.35
22
GPT-5.6 Luna Pro (batch)Budget
OpenAI$0.10$0.60$0.35
23
GPT-5.4 Nano (batch)Budget
OpenAI$0.10$0.63$0.36
24
DeepSeek V4.1 FlashBudget
DeepSeek$0.02$0.72$0.37
25
Mistral Small 4Budget
Mistral$0.15$0.60$0.38
26
DeepSeek V4 Flash Vision ExpBudget
DeepSeek$0.22$0.65$0.43
27
Gemini 3.1 Flash Lite (batch)Budget
Google$0.13$0.75$0.44
28
Qwen3 Coder NextBudget
Qwen$0.12$0.80$0.46
29
Qwen3.5-35B-A3BBudget
Qwen$0.15$1.00$0.57
30
Qwen3.6 35B A3BBudget
Qwen$0.15$1.00$0.57
31
Qwen3.6 FlashBudget
Qwen$0.19$1.13$0.66
32
Gemini 3.5 Flash Lite (batch)Budget
Google$0.15$1.25$0.70
33
GPT-5.6 LunaBudget
OpenAI$0.20$1.20$0.70
34
GPT-5.6 Luna ProBudget
OpenAI$0.20$1.20$0.70
35
GPT-5.4 NanoBudget
OpenAI$0.20$1.25$0.72
36
Qwen3.7 PlusBudget
Qwen$0.32$1.28$0.80
37
GLM 5.3 FlashXBudget
Z.AI$0.37$1.25$0.81
38
Gemini 3.1 Flash LiteBudget
Google$0.25$1.50$0.88
39
Gemini 3.1 Flash Lite PreviewBudget
Google$0.25$1.50$0.88
40
Command A+Budget
Cohere$0.30$1.50$0.90
41
Qwen3.5 Plus 2026-02-15Budget
Qwen$0.26$1.56$0.91
42
Qwen3.5 Plus 2026-04-20Budget
Qwen$0.30$1.80$1.05
43
Gemini 3.6 Flash (batch)Budget
Google$0.38$1.88$1.13
44
Gemini 3.7 Flash (batch)Budget
Google$0.38$1.88$1.13
45
Gemini 3.8 Flash (batch)Budget
Google$0.38$1.88$1.13
46
Qwen3.6 PlusBudget
Qwen$0.33$1.95$1.14
47
Qwen3.5-122B-A10BBudget
Qwen$0.26$2.08$1.17
48
Qwen3.8 27BBudget
Qwen$0.24$2.20$1.22
49
GLM 5.3 (batch)Budget
Z.AI$0.45$2.00$1.23
50
GPT-5.4 Mini (batch)Budget
OpenAI$0.38$2.25$1.31
51
DeepSeek V4 Pro 0813Budget
DeepSeek$0.66$1.98$1.32
52
Mistral Large 4Budget
Mistral$0.68$2.09$1.39
53
Gemini 3.5 Flash LiteBudget
Google$0.30$2.50$1.40
54
Qwen3.5-27BBudget
Qwen$0.26$2.60$1.43
55
DeepSeek V4 Pro 0423Budget
DeepSeek$0.95$1.91$1.43
56
Kimi K2.5Budget
Moonshot$0.50$2.50$1.50
57
Grok 4.3 (batch)Standard
xAI$1.00$2.00$1.50
58
Grok Build 0.1Standard
xAI$1.00$2.00$1.50
59
Qwen3.6 27BBudget
Qwen$0.32$3.20$1.76
60
Grok 4.20Standard
xAI$1.25$2.50$1.88
61
Grok 4.20 Multi-AgentStandard
xAI$1.25$2.50$1.88
62
Grok 4.3Standard
xAI$1.25$2.50$1.88
63
GLM 5.1Budget
Z.AI$0.97$3.04$2.00
64
Kimi K2.7 CodeBudget
Moonshot$0.67$3.35$2.01
65
Kimi K2.6Budget
Moonshot$0.65$3.41$2.03
66
Qwen3.5 397B A17BBudget
Qwen$0.50$3.60$2.05
67
GLM 5Standard
Z.AI$1.00$3.20$2.10
68
Gemini 3.6 FlashBudget
Google$0.75$3.75$2.25
69
Gemini 3.7 FlashBudget
Google$0.75$3.75$2.25
70
Gemini 3.8 FlashBudget
Google$0.75$3.75$2.25
71
Mistral Medium 3.5 (batch)Budget
Mistral$0.75$3.75$2.25
72
GLM 5.3Budget
Z.AI$0.03$4.80$2.42
73
GLM 5 TurboStandard
Z.AI$1.20$4.00$2.60
74
GLM 5V TurboStandard
Z.AI$1.20$4.00$2.60
75
Gemini 3.5 Flash (batch)Budget
Google$0.75$4.50$2.63
76
GPT-5.4 MiniBudget
OpenAI$0.75$4.50$2.63
77
Qwen3.7 MaxStandard
Qwen$1.48$4.42$2.95
78
Claude Sonnet 5 (batch)Standard
Anthropic$1.00$5.00$3.00
79
Claude Sonnet 5.5 (batch)Standard
Anthropic$1.00$5.00$3.00
80
Gemini 4 Argon (batch)Standard
Google$1.00$5.00$3.00
81
GPT-6 Sol (batch)Standard
OpenAI$1.00$5.00$3.00
82
GPT-6 Sol Pro (batch)Standard
OpenAI$1.00$5.00$3.00
83
GPT-6.1 Sol (batch)Standard
OpenAI$1.00$5.00$3.00
84
GPT-6.1 Sol Pro (batch)Standard
OpenAI$1.00$5.00$3.00
85
Gemini 3.1 Pro Preview (batch)Standard
Google$1.00$6.00$3.50
86
GPT-5.6 Terra (batch)Standard
OpenAI$1.00$6.00$3.50
87
GPT-5.6 Terra Pro (batch)Standard
OpenAI$1.00$6.00$3.50
88
GLM 5.2Budget
Z.AI$0.02$7.00$3.51
89
Qwen3.8 2.4T A95BStandard
Qwen$2.00$6.00$4.00
90
Qwen3.8 Max (0902)Standard
Qwen$2.00$6.00$4.00
91
Grok 4.5Standard
xAI$2.00$6.00$4.00
92
Grok 4.6Standard
xAI$2.00$6.00$4.00
93
Grok 4.7Standard
xAI$2.00$6.00$4.00
94
GPT-5.4 (batch)Standard
OpenAI$1.25$7.50$4.38
95
Claude Sonnet 4.6 (batch)Standard
Anthropic$1.50$7.50$4.50
96
Nano Banana 2.1Standard
Google$1.50$7.50$4.50
97
Mistral Medium 3.5Standard
Mistral$1.50$7.50$4.50
98
Gemini 3.5 FlashStandard
Google$1.50$9.00$5.25
99
GLM 5.3 PrimeStandard
Z.AI$2.80$8.80$5.80
100
Claude Opus 5.5 (batch)Standard
Anthropic$2.00$10.00$6.00
101
Claude Sonnet 5Standard
Anthropic$2.00$10.00$6.00
102
Claude Sonnet 5.5Standard
Anthropic$2.00$10.00$6.00
103
Gemini 4 ArgonStandard
Google$2.00$10.00$6.00
104
GPT-5.6 Sol (batch)Standard
OpenAI$2.00$10.00$6.00
105
GPT-5.6 Sol Pro (batch)Standard
OpenAI$2.00$10.00$6.00
106
GPT-6 SolStandard
OpenAI$2.00$10.00$6.00
107
GPT-6 Sol ProStandard
OpenAI$2.00$10.00$6.00
108
GPT-6.1 SolStandard
OpenAI$2.00$10.00$6.00
109
GPT-6.1 Sol ProStandard
OpenAI$2.00$10.00$6.00
110
Kimi K3 (batch)Standard
Moonshot$2.28$11.40$6.84
111
Gemini 3.1 Pro PreviewStandard
Google$2.00$12.00$7.00
112
Gemini 3.1 Pro Preview Custom ToolsStandard
Google$2.00$12.00$7.00
113
GPT-5.6 TerraStandard
OpenAI$2.00$12.00$7.00
114
GPT-5.6 Terra ProStandard
OpenAI$2.00$12.00$7.00
115
Claude Opus 4.6 (batch)Standard
Anthropic$2.50$12.50$7.50
116
Claude Opus 4.7 (batch)Standard
Anthropic$2.50$12.50$7.50
117
Claude Opus 4.8 (batch)Standard
Anthropic$2.50$12.50$7.50
118
Claude Opus 5 (batch)Standard
Anthropic$2.50$12.50$7.50
119
Kimi K3Budget
Moonshot$0.45$15.00$7.72
120
GPT-5.2-CodexStandard
OpenAI$1.75$14.00$7.88
121
GPT-5.3-CodexStandard
OpenAI$1.75$14.00$7.88
122
Qwen3.8 Max PrimeStandard
Qwen$4.00$12.00$8.00
123
GPT-5.4Standard
OpenAI$2.50$15.00$8.75
124
GPT-5.5 (batch)Standard
OpenAI$2.50$15.00$8.75
125
Claude Sonnet 4.6Standard
Anthropic$3.00$15.00$9.00
126
Claude Opus 5.5Standard
Anthropic$4.00$20.00$12.00
127
GPT-5.6 SolStandard
OpenAI$4.00$20.00$12.00
128
GPT-5.6 Sol ProStandard
OpenAI$4.00$20.00$12.00
129
Claude Fable 5 (batch)Premium
Anthropic$5.00$25.00$15.00
130
Claude Fable 5.1 (batch)Premium
Anthropic$5.00$25.00$15.00
131
Claude Opus 4.6Premium
Anthropic$5.00$25.00$15.00
132
Claude Opus 4.7Premium
Anthropic$5.00$25.00$15.00
133
Claude Opus 4.8Premium
Anthropic$5.00$25.00$15.00
134
Claude Opus 5Premium
Anthropic$5.00$25.00$15.00
135
GPT-6 Astra (batch)Premium
OpenAI$5.00$25.00$15.00
136
GPT-6 Astra Pro (batch)Premium
OpenAI$5.00$25.00$15.00
137
GPT Chat LatestPremium
OpenAI$5.00$30.00$17.50
138
GPT-5.5Premium
OpenAI$5.00$30.00$17.50
139
Claude Fable 5Premium
Anthropic$10.00$50.00$30.00
140
Claude Fable 5.1Premium
Anthropic$10.00$50.00$30.00
141
GPT-6 AstraPremium
OpenAI$10.00$50.00$30.00
142
GPT-6 Astra ProPremium
OpenAI$10.00$50.00$30.00
143
GPT-5.4 Pro (batch)Premium
OpenAI$15.00$90.00$52.50
144
GPT-5.5 Pro (batch)Premium
OpenAI$15.00$90.00$52.50
145
GPT-5.4 ProPremium
OpenAI$30.00$180.00$105.00
146
GPT-5.5 ProPremium
OpenAI$30.00$180.00$105.00

Blended price = input rate × input share + output rate × output share, per 1M tokens. Refreshed daily. Full input/output/cached rates are on the pricing page.

How to find the cheapest model for your use case

The lowest headline price isn’t always the cheapest model to run. Because input and output tokens are priced separately — and output usually costs several times more — the winner flips depending on your workload:

  • Input-heavy (80/20) — retrieval-augmented generation, classification, extraction, routing. You send a lot and generate a little, so the input rate dominates.
  • Output-heavy (20/80) — drafting, long-form writing, code generation. The output rate dominates, so a low output price matters most.
  • Balanced (50/50) — chat and general assistants, a reasonable default when you’re unsure.

The blended figure reweights each model’s input and output rates for the mix you pick, so the ranking reflects what you’ll really pay. Once you’ve shortlisted, sanity-check capability against price on the benchmarks — the very cheapest model isn’t a bargain if it can’t do the job — and project a full monthly bill on the calculator. To price a specific prompt, use the token counter. For the full breakdown of why the ranking depends on your ratio, see Which Frontier Model Is Actually Cheapest for Your Workload? — and for ways to cut cost without switching models at all, see How to Cut Your LLM API Bill by 80%.

Frequently asked questions

What is the cheapest LLM API right now?
It depends on your workload. For output-heavy tasks, the model with the lowest output price usually wins; for input-heavy tasks, the lowest input price wins. Use the workload toggle above to rank models by the blended price for your mix — the top row is the cheapest for that scenario, refreshed daily.
What does 'blended price' mean?
A single per-1M-token figure that combines a model’s input and output rates weighted by your workload mix. For a 50/50 balanced mix it is the average of the input and output rates; for an 80/20 input-heavy mix it leans toward the input rate.
Is the cheapest model always the best choice?
No. A cheap model that can’t complete your task reliably costs more in retries and engineering time. Weigh price against capability on the benchmarks page, and validate on your own tasks before committing.
How often do the rankings update?
Prices refresh automatically every day from live provider data, so the cheapest-first ranking reflects current rates. Always confirm the final price on the provider’s own page before committing spend.