Tool · 2026-06-29
LLM‑Preisrechner
Geschätzte monatliche Kosten nach Ihrem Token‑Volumen.
Total tokens / month: 465,000,000 (360,000,000 in + 105,000,000 out)
| # | Model | Vendor | In $/1M | Out $/1M | Monthly cost | vs cheapest |
|---|---|---|---|---|---|---|
| 1 | Meta-Llama-3.1-8B-Instruct | Meta | $0.020 | $0.030 | $10.35 | — |
| 2 | gpt-oss-20b | OpenAI | $0.029 | $0.140 | $25.14 | 2.4× |
| 3 | gpt-oss-120b | OpenAI | $0.030 | $0.150 | $26.55 | 2.6× |
| 4 | Llama-3.3-70B-Instruct | Meta | $0.050 | $0.230 | $42.15 | 4.1× |
| 5 | GLM-4.7-Flash | Z.AI / Zhipu | $0.040 | $0.300 | $45.90 | 4.4× |
| 6 | Qwen3 235B A22B Instruct 2507 | Alibaba (Qwen) | $0.100 | $0.100 | $46.50 | 4.5× |
| 7 | Qwen3-235B-A22B-Thinking-2507 | Alibaba (Qwen) | $0.100 | $0.100 | $46.50 | 4.5× |
| 8 | Gemma 4 26B A4B IT | $0.060 | $0.330 | $56.25 | 5.4× | |
| 9 | GPT-5 Nano | OpenAI | $0.050 | $0.400 | $60.00 | 5.8× |
| 10 | Gemma 4 31B IT | $0.100 | $0.300 | $67.50 | 6.5× | |
| 11 | Gemini 2.5 Flash-Lite | $0.100 | $0.400 | $78.00 | 7.5× | |
| 12 | GPT-4.1 nano | OpenAI | $0.100 | $0.400 | $78.00 | 7.5× |
| 13 | DeepSeek V4 Flash | DeepSeek | $0.140 | $0.280 | $79.80 | 7.7× |
| 14 | DeepSeek-V3.2 | DeepSeek | $0.180 | $0.350 | $101.55 | 9.8× |
| 15 | Qwen3 Coder Next | Alibaba (Qwen) | $0.108 | $0.675 | $109.76 | 10.6× |
| 16 | GPT-4o mini | OpenAI | $0.150 | $0.600 | $117.00 | 11.3× |
| 17 | DeepSeek-V3.1 | DeepSeek | $0.200 | $0.700 | $145.50 | 14.1× |
| 18 | GLM-4.5-Air | Z.AI / Zhipu | $0.200 | $1.10 | $187.50 | 18.1× |
| 19 | GPT-5.4 nano | OpenAI | $0.200 | $1.25 | $203.25 | 19.6× |
| 20 | MiniMax-M2.5 | MiniMax | $0.300 | $1.20 | $234.00 | 22.6× |
| 21 | MiniMax-M2.7 | MiniMax | $0.300 | $1.20 | $234.00 | 22.6× |
| 22 | MiniMax-M2.1 | MiniMax | $0.300 | $1.20 | $234.00 | 22.6× |
| 23 | MiniMax-M3 | MiniMax | $0.300 | $1.20 | $234.00 | 22.6× |
| 24 | MiniMax-M2 | MiniMax | $0.300 | $1.20 | $234.00 | 22.6× |
| 25 | Qwen3.6 35B-A3B | Alibaba (Qwen) | $0.248 | $1.49 | $245.21 | 23.7× |
| 26 | DeepSeek V4 Pro | DeepSeek | $0.435 | $0.870 | $247.95 | 24.0× |
| 27 | GPT-5 Mini | OpenAI | $0.250 | $2.00 | $300.00 | 29.0× |
| 28 | GPT-5.1 Codex mini | OpenAI | $0.250 | $2.00 | $300.00 | 29.0× |
| 29 | GPT-4.1 mini | OpenAI | $0.400 | $1.60 | $312.00 | 30.1× |
| 30 | DeepSeek-R1 | DeepSeek | $0.400 | $1.70 | $322.50 | 31.2× |
| 31 | DeepSeek-R1-0528 | DeepSeek | $0.400 | $1.70 | $322.50 | 31.2× |
| 32 | Gemini 2.5 Flash | $0.300 | $2.50 | $370.50 | 35.8× | |
| 33 | Qwen3-Next 80B-A3B Instruct | Alibaba (Qwen) | $0.500 | $2.00 | $390.00 | 37.7× |
| 34 | Qwen3-Coder 30B-A3B Instruct | Alibaba (Qwen) | $0.450 | $2.25 | $398.25 | 38.5× |
| 35 | GLM-4.7 | Z.AI / Zhipu | $0.600 | $2.20 | $447.00 | 43.2× |
| 36 | GLM-4.6 | Z.AI / Zhipu | $0.600 | $2.20 | $447.00 | 43.2× |
| 37 | GLM-4.5 | Z.AI / Zhipu | $0.600 | $2.20 | $447.00 | 43.2× |
| 38 | Kimi K2 Thinking | Moonshot AI | $0.600 | $2.50 | $478.50 | 46.2× |
| 39 | Gemini 3 Flash Preview | $0.500 | $3.00 | $495.00 | 47.8× | |
| 40 | Qwen3.6 Plus | Alibaba (Qwen) | $0.500 | $3.00 | $495.00 | 47.8× |
| 41 | Kimi K2.5 | Moonshot AI | $0.600 | $3.00 | $531.00 | 51.3× |
| 42 | Qwen3 32B | Alibaba (Qwen) | $0.700 | $2.80 | $546.00 | 52.8× |
| 43 | Qwen3.5 397B-A17B | Alibaba (Qwen) | $0.600 | $3.60 | $594.00 | 57.4× |
| 44 | GLM-5 | Z.AI / Zhipu | $1.00 | $3.20 | $696.00 | 67.2× |
| 45 | Grok 4.3 | xAI | $1.25 | $2.50 | $712.50 | 68.8× |
| 46 | GPT-5.4 mini | OpenAI | $0.750 | $4.50 | $742.50 | 71.7× |
| 47 | Kimi K2.6 | Moonshot AI | $0.950 | $4.00 | $762.00 | 73.6× |
| 48 | Kimi K2.7 Code | Moonshot AI | $0.950 | $4.00 | $762.00 | 73.6× |
| 49 | o3-mini | OpenAI | $1.10 | $4.40 | $858.00 | 82.9× |
| 50 | Claude Haiku 4.5 (latest) | Anthropic | $1.00 | $5.00 | $885.00 | 85.5× |
| 51 | GLM-5.1 | Z.AI / Zhipu | $1.40 | $4.40 | $966.00 | 93.3× |
| 52 | GLM-5.2 | Z.AI / Zhipu | $1.40 | $4.40 | $966.00 | 93.3× |
| 53 | Qwen3-Coder 480B-A35B Instruct | Alibaba (Qwen) | $1.50 | $7.50 | $1,328 | 128.3× |
| 54 | Gemini 3.5 Flash | $1.50 | $9.00 | $1,485 | 143.5× | |
| 55 | Gemini 2.5 Pro | $1.25 | $10.00 | $1,500 | 144.9× | |
| 56 | GPT-5 | OpenAI | $1.25 | $10.00 | $1,500 | 144.9× |
| 57 | GPT-5.1 | OpenAI | $1.25 | $10.00 | $1,500 | 144.9× |
| 58 | GPT-5.1 Codex | OpenAI | $1.25 | $10.00 | $1,500 | 144.9× |
| 59 | GPT-5-Codex | OpenAI | $1.25 | $10.00 | $1,500 | 144.9× |
| 60 | GPT-5.1 Codex Max | OpenAI | $1.25 | $10.00 | $1,500 | 144.9× |
| 61 | GPT-4.1 | OpenAI | $2.00 | $8.00 | $1,560 | 150.7× |
| 62 | o3 | OpenAI | $2.00 | $8.00 | $1,560 | 150.7× |
| 63 | Qwen3.7 Max | Alibaba (Qwen) | $2.50 | $7.50 | $1,688 | 163.0× |
| 64 | GPT-4o | OpenAI | $2.50 | $10.00 | $1,950 | 188.4× |
| 65 | Gemini 3.1 Pro Preview | $2.00 | $12.00 | $1,980 | 191.3× | |
| 66 | GPT-5.2 | OpenAI | $1.75 | $14.00 | $2,100 | 202.9× |
| 67 | GPT-5.3 Codex | OpenAI | $1.75 | $14.00 | $2,100 | 202.9× |
| 68 | GPT-5.2 Codex | OpenAI | $1.75 | $14.00 | $2,100 | 202.9× |
| 69 | Claude Sonnet 4 (latest) | Anthropic | $2.60 | $13.00 | $2,301 | 222.3× |
| 70 | GPT-5.4 | OpenAI | $2.50 | $15.00 | $2,475 | 239.1× |
| 71 | Claude Sonnet 4.6 | Anthropic | $3.00 | $15.00 | $2,655 | 256.5× |
| 72 | Claude Sonnet 4.5 (latest) | Anthropic | $3.00 | $15.00 | $2,655 | 256.5× |
| 73 | Claude Opus 4.6 | Anthropic | $5.00 | $25.00 | $4,425 | 427.5× |
| 74 | Claude Opus 4.7 | Anthropic | $5.00 | $25.00 | $4,425 | 427.5× |
| 75 | Claude Opus 4.8 | Anthropic | $5.00 | $25.00 | $4,425 | 427.5× |
| 76 | Claude Opus 4.5 (latest) | Anthropic | $5.00 | $25.00 | $4,425 | 427.5× |
| 77 | GPT-5.5 | OpenAI | $5.00 | $30.00 | $4,950 | 478.3× |
| 78 | Claude Opus 4.1 (latest) | Anthropic | $15.00 | $75.00 | $13,275 | 1282.6× |
| 79 | GPT-5 Pro | OpenAI | $15.00 | $120.00 | $18,000 | 1739.1× |
| 80 | GPT-5.4 Pro | OpenAI | $30.00 | $180.00 | $29,700 | 2869.6× |
So wird gerechnet
monthly_cost = requests × ((avg_input_tokens × input_price + avg_output_tokens × output_price) / 1,000,000).
Numbers shown are estimates only. Real-world cost depends on prompt caching, >200K context tier rates, audio/image surcharges and provider-specific overage. Always confirm with the model detail page or the provider's official pricing.
Frequently asked questions
How accurate are the cost estimates?
We multiply your average input/output token counts by the recommended provider's headline rate from our daily-refreshed canonical catalogue. The math is exact, but the result is only as accurate as your token-count assumptions. Real production cost will also depend on prompt caching, batch-API discounts, >200K-context tier rates and audio/image surcharges — none of which the calculator factors in.
Where do the per-token prices come from?
All prices reflect each provider's published list rate in USD, refreshed daily from our normalised catalogue. They are NOT including taxes, prepaid credits or volume discounts.
Can I share or bookmark a calculation?
Yes. The URL captures your selected model and token settings, so you can link a teammate to a specific scenario without retyping the inputs.
Why are some models missing from the picker?
We only include text-LLMs with a published per-token price. Models without published rates (often invite-only or enterprise-gated) and embedding/audio-only models are excluded — the calculator is text-completion-focused.
How do I estimate cost for caching or long context?
Open the model's detail page — the Pricing detail block lists cache_read / cache_write / context_over_200k rates where applicable. The headline calculator number is a 'no-cache, ≤200K' baseline.
Explore more
Other tools
Pricing comparisons
Best-of model lists
Browse by capability
Background reading
Prices in USD per 1M tokens. Unknown means the provider does not publish per-token pricing.
Pricing and capabilities are refreshed daily and reconciled against each provider's official documentation. Always verify critical production decisions with the provider directly.