Documents
A 100-page report sent whole to the model, answered with a one-page executive summary.
Pricing verified:
Between $0.0036 and $0.690 per run, depending on the model. That is a 193× spread for the same job.
100 pages of dense prose at ~500 words per page = ~50,000 words ≈ 65,000 input tokens. The summary back is ~600 words ≈ 800 output tokens.
Cost = (input tokens ÷ 1M × input price) + (output tokens ÷ 1M × output price). Prices come from our daily-verified model data. Batch and cached-input discounts are not applied — they only ever make these numbers smaller.
| Model | Provider | Per run | At 100 PDFs a month | Context |
|---|---|---|---|---|
| Llama 3.2 1B Instructneeds chunking | Meta | $0.0019 | $0.192 | 60k tokens |
| Gemma 2 9Bneeds chunking | $0.0020 | $0.202 | 8k tokens | |
| Llama 3.1 8B Instructneeds chunking | Meta | $0.0033 | $0.331 | 16k tokens |
| GPT-5 Nano | OpenAI | $0.0036 | $0.357 | 400k tokens |
| gpt-oss-safeguard-20b | OpenAI | $0.0047 | $0.471 | 131k tokens |
| Mistral Small 3.2 24B | Mistral | $0.0050 | $0.503 | 128k tokens |
| Gemini 2.0 Flash Lite | $0.0051 | $0.511 | 1.0M tokens | |
| Ministral 3 3B 2512 | Mistral | $0.0066 | $0.658 | 131k tokens |
| Devstral Small 1.1 | Mistral | $0.0067 | $0.674 | 131k tokens |
| Mistral Small Creativeneeds chunking | Mistral | $0.0067 | $0.674 | 33k tokens |
| Voxtral Small 24B 2507needs chunking | Mistral | $0.0067 | $0.674 | 32k tokens |
| Mistral Small 3.1 | Mistral | $0.0067 | $0.674 | 128k tokens |
| Gemini 2.0 Flash | $0.0068 | $0.682 | 1.0M tokens | |
| Gemini 2.5 Flash Lite | $0.0068 | $0.682 | 1.0M tokens | |
| Gemini 2.5 Flash Lite Preview 09-2025 | $0.0068 | $0.682 | 1.0M tokens | |
| GPT-4.1 Nano | OpenAI | $0.0068 | $0.682 | 1.0M tokens |
| Mistral 7B Instruct v0.1needs chunking | Mistral | $0.0073 | $0.730 | 3k tokens |
| Gemma 4 26B A4B | $0.0088 | $0.877 | 262k tokens | |
| Llama 3 8B Instructneeds chunking | Meta | $0.0092 | $0.921 | 8k tokens |
| DeepSeek V4-Flash | DeepSeek | $0.0093 | $0.932 | 1M tokens |
| Ministral 3 8B 2512 | Mistral | $0.0099 | $0.987 | 262k tokens |
| Mistral Nemo | Mistral | $0.0099 | $0.987 | 131k tokens |
| GLM-5.3 Flash | Z.ai | $0.010 | $1.01 | 1M tokens |
| GPT-4o Mini | OpenAI | $0.010 | $1.02 | 128k tokens |
| Mistral Small 4 | Mistral | $0.010 | $1.02 | 262k tokens |
| Qwen 3.8 Flash | Alibaba | $0.011 | $1.08 | 991k tokens |
| Llama Guard 4 12B | Meta | $0.012 | $1.18 | 164k tokens |
| Ministral 3 14B 2512 | Mistral | $0.013 | $1.32 | 262k tokens |
| Sabaneeds chunking | Mistral | $0.013 | $1.35 | 33k tokens |
| GPT-5.6 Luna | OpenAI | $0.014 | $1.40 | 1.1M tokens |
| Grok Code Fast 1 | xAI | $0.014 | $1.42 | 256k tokens |
| Claude 3 Haiku | Anthropic | $0.017 | $1.73 | 200k tokens |
| Gemini 3 Flash Preview | $0.017 | $1.75 | 1.0M tokens | |
| GPT-5 Mini | OpenAI | $0.018 | $1.79 | 400k tokens |
| GPT-5.1-Codex-Mini | OpenAI | $0.018 | $1.79 | 400k tokens |
| DeepSeek V3 | DeepSeek | $0.018 | $1.84 | 128k tokens |
| Grok 3 Mini | xAI | $0.020 | $1.99 | 131k tokens |
| Grok 3 Mini Beta | xAI | $0.020 | $1.99 | 131k tokens |
| Codestral 2508 | Mistral | $0.020 | $2.02 | 256k tokens |
| Gemini 2.5 Flash | $0.021 | $2.15 | 1.0M tokens | |
| Gemini 3.5 Flash-Lite | $0.021 | $2.15 | 1.0M tokens | |
| Nano Banana (Gemini 2.5 Flash Image)needs chunking | $0.021 | $2.15 | 33k tokens | |
| Llama 3.2 11B Vision Instruct | Meta | $0.023 | $2.27 | 131k tokens |
| Mistral Small 3needs chunking | Mistral | $0.023 | $2.33 | 33k tokens |
| Muse Glimmer 30B | Meta | $0.024 | $2.40 | 131k tokens |
| Gemma 4 31B | $0.026 | $2.61 | 262k tokens | |
| Llama 3.1 70B Instruct | Meta | $0.026 | $2.63 | 131k tokens |
| GPT-4.1 Mini | OpenAI | $0.027 | $2.73 | 1.0M tokens |
| Devstral 2 2512 | Mistral | $0.028 | $2.76 | 262k tokens |
| Devstral Medium | Mistral | $0.028 | $2.76 | 131k tokens |
| Mistral Medium 3 | Mistral | $0.028 | $2.76 | 131k tokens |
| Mistral Medium 3.1 | Mistral | $0.028 | $2.76 | 131k tokens |
| DeepSeek V4-Pro | DeepSeek | $0.029 | $2.90 | 1M tokens |
| Llama Guard 3 8B | Meta | $0.031 | $3.12 | 131k tokens |
| Llama 4 Scout | Meta | $0.033 | $3.35 | 512k tokens |
| GPT-3.5 Turboneeds chunking | OpenAI | $0.034 | $3.37 | 16k tokens |
| Mistral Large 3 2512 | Mistral | $0.034 | $3.37 | 262k tokens |
| Llama 3 70B Instructneeds chunking | Meta | $0.034 | $3.37 | 8k tokens |
| Gemini 3.1 Flash | $0.035 | $3.49 | 1M tokens | |
| Mixtral 8x7B Instructneeds chunking | Mistral | $0.036 | $3.55 | 33k tokens |
| DeepSeek R1 | DeepSeek | $0.038 | $3.75 | 128k tokens |
| Llama 4 Maverick | Meta | $0.040 | $4.03 | 256k tokens |
| GPT Audio Mini | OpenAI | $0.041 | $4.09 | 128k tokens |
| Gemma 2 27Bneeds chunking | $0.043 | $4.28 | 8k tokens | |
| Gemini 3.6 Flash | $0.052 | $5.18 | 1.0M tokens | |
| Gemini 3.7 Flash | $0.052 | $5.18 | 1.0M tokens | |
| Claude 3.5 Haiku | Anthropic | $0.055 | $5.52 | 200k tokens |
| Claude 4 Haiku | Anthropic | $0.055 | $5.52 | 200k tokens |
| Codestral 25.01 | Mistral | $0.061 | $6.07 | 256k tokens |
| Kimi K2.7 Code | Moonshot | $0.065 | $6.49 | 256k tokens |
| GPT-3.5 Turbo (older v0613)needs chunking | OpenAI | $0.067 | $6.66 | 4k tokens |
| Claude Haiku 4.5 | Anthropic | $0.069 | $6.90 | 200k tokens |
| o3 Mini | OpenAI | $0.075 | $7.50 | 200k tokens |
| o3 Mini High | OpenAI | $0.075 | $7.50 | 200k tokens |
| o4 Mini | OpenAI | $0.075 | $7.50 | 200k tokens |
| o4 Mini High | OpenAI | $0.075 | $7.50 | 200k tokens |
| GPT-5.2 Mini | OpenAI | $0.082 | $8.18 | 128k tokens |
| Muse Spark | Meta | $0.085 | $8.46 | 1.0M tokens |
| GPT-5 | OpenAI | $0.089 | $8.92 | 400k tokens |
| GPT-5 Chat | OpenAI | $0.089 | $8.92 | 128k tokens |
| GPT-5 Codex | OpenAI | $0.089 | $8.92 | 400k tokens |
| GPT-5.1 | OpenAI | $0.089 | $8.92 | 400k tokens |
| GPT-5.1 Chat | OpenAI | $0.089 | $8.92 | 128k tokens |
| GPT-5.1-Codex | OpenAI | $0.089 | $8.92 | 400k tokens |
| GPT-5.1-Codex-Max | OpenAI | $0.089 | $8.92 | 400k tokens |
| Gemini 2.5 Pro | $0.089 | $8.92 | 1.0M tokens | |
| Gemini 2.5 Pro Preview 05-06 | $0.089 | $8.92 | 1.0M tokens | |
| Gemini 2.5 Pro Preview 06-05 | $0.089 | $8.92 | 1.0M tokens | |
| GLM-5.2 | Z.ai | $0.095 | $9.45 | 1M tokens |
| GLM-5.3 | Z.ai | $0.095 | $9.45 | 1M tokens |
| GPT-3.5 Turbo Instructneeds chunking | OpenAI | $0.099 | $9.91 | 4k tokens |
| Mistral Medium 3.5 | Mistral | $0.104 | $10.35 | 256k tokens |
| Gemini 3.5 Flash | $0.105 | $10.47 | 1.0M tokens | |
| GPT-5.2 | OpenAI | $0.125 | $12.50 | 200k tokens |
| GPT-5.3 Chat | OpenAI | $0.125 | $12.50 | 128k tokens |
| GPT-5.3-Codex | OpenAI | $0.125 | $12.50 | 400k tokens |
| Grok 4 | xAI | $0.135 | $13.48 | 2M tokens |
| Grok 4.5 | xAI | $0.135 | $13.48 | 500k tokens |
| Grok 4.6 | xAI | $0.135 | $13.48 | 500k tokens |
| Mixtral 8x22B Instructneeds chunking | Mistral | $0.135 | $13.48 | 66k tokens |
| Pixtral Large 2411 | Mistral | $0.135 | $13.48 | 131k tokens |
| Qwen 3.8 Max | Alibaba | $0.135 | $13.48 | 1M tokens |
| GPT-4.1 | OpenAI | $0.136 | $13.64 | 1.0M tokens |
| o3 | OpenAI | $0.136 | $13.64 | 200k tokens |
| o4 Mini Deep Research | OpenAI | $0.136 | $13.64 | 200k tokens |
| Claude Sonnet 5 | Anthropic | $0.138 | $13.80 | 1M tokens |
| GPT-5.6 Sol | OpenAI | $0.138 | $13.80 | 1.1M tokens |
| GPT-5.6 Terra | OpenAI | $0.140 | $13.96 | 1.1M tokens |
| Gemini 3.1 Pro | $0.140 | $13.96 | 2M tokens | |
| Nano Banana Pro (Gemini 3 Pro Image Preview)needs chunking | $0.140 | $13.96 | 66k tokens | |
| GPT-5 Image Mini | OpenAI | $0.164 | $16.41 | 400k tokens |
| Qwen 3.7 Max | Alibaba | $0.169 | $16.85 | 1M tokens |
| GPT Audio | OpenAI | $0.171 | $17.05 | 128k tokens |
| GPT-4o | OpenAI | $0.171 | $17.05 | 128k tokens |
| GPT-5.4 | OpenAI | $0.175 | $17.45 | 272k tokens |
| GPT-3.5 Turbo 16kneeds chunking | OpenAI | $0.198 | $19.82 | 16k tokens |
| Mistral Large 2 | Mistral | $0.202 | $20.22 | 128k tokens |
| Claude 3.7 Sonnet (thinking) | Anthropic | $0.207 | $20.70 | 200k tokens |
| Claude Sonnet 4 | Anthropic | $0.207 | $20.70 | 200k tokens |
| Claude Sonnet 4.5 | Anthropic | $0.207 | $20.70 | 1M tokens |
| Claude Sonnet 4.6 | Anthropic | $0.207 | $20.70 | 1M tokens |
| Grok 3 | xAI | $0.207 | $20.70 | 131k tokens |
| Grok 3 Beta | xAI | $0.207 | $20.70 | 131k tokens |
| Kimi K3 | Moonshot | $0.207 | $20.70 | 1M tokens |
| Claude Opus 4.5 | Anthropic | $0.345 | $34.50 | 200k tokens |
| Claude Opus 4.6 | Anthropic | $0.345 | $34.50 | 1M tokens |
| Claude Opus 4.7 | Anthropic | $0.345 | $34.50 | 1M tokens |
| Claude Opus 4.8 | Anthropic | $0.345 | $34.50 | 1M tokens |
| Claude Opus 5 | Anthropic | $0.345 | $34.50 | 1M tokens |
| GPT-5.5 | OpenAI | $0.349 | $34.90 | 1M tokens |
| Claude 3.5 Sonnet | Anthropic | $0.414 | $41.40 | 200k tokens |
| GPT-5 Image | OpenAI | $0.658 | $65.80 | 400k tokens |
| GPT-4 Turbo | OpenAI | $0.674 | $67.40 | 128k tokens |
| GPT-4 Turbo (older v1106) | OpenAI | $0.674 | $67.40 | 128k tokens |
| GPT-4 Turbo Preview | OpenAI | $0.674 | $67.40 | 128k tokens |
| o3 Deep Research | OpenAI | $0.682 | $68.20 | 200k tokens |
| Claude Fable 5 | Anthropic | $0.690 | $69.00 | 1M tokens |
| Claude Fable 5.1 | Anthropic | $0.690 | $69.00 | 1M tokens |
| Claude Mythos 5 | Anthropic | $0.690 | $69.00 | 1M tokens |
| Claude Mythos 5.1 | Anthropic | $0.690 | $69.00 | 1M tokens |
| GPT-6 Astra | OpenAI | $0.690 | $69.00 | 1.1M tokens |
| o1 | OpenAI | $1.02 | $102 | 200k tokens |
| Claude Opus 4 | Anthropic | $1.04 | $104 | 200k tokens |
| Claude Opus 4.1 | Anthropic | $1.04 | $104 | 200k tokens |
| GPT-5 Pro | OpenAI | $1.07 | $107 | 400k tokens |
| o3 Pro | OpenAI | $1.36 | $136 | 200k tokens |
| GPT-4needs chunking | OpenAI | $2.00 | $200 | 8k tokens |
| GPT-4 (older v0314)needs chunking | OpenAI | $2.00 | $200 | 8k tokens |
| o1-pro | OpenAI | $10.23 | $1,023 | 200k tokens |
“Needs chunking” means the model's context window can't hold this job in a single pass, so the real cost is higher than the figure shown and quality usually suffers.
This is an input-heavy job, so the input price per million tokens decides almost the whole bill — the summary you get back is rounding error. Check the context window before the price: a model that can't hold 65,000 tokens has to be fed in chunks, which costs more and summarises worse.
$0.0036 on GPT-5 Nano, the cheapest capable option, rising to $0.690 on GPT-6 Astra at the top end. GPT-5 Nano is the value pick at $0.0036 per run. The job is priced at 65,000 input and 800 output tokens — see the working below.
100 pages of dense prose at ~500 words per page = ~50,000 words ≈ 65,000 input tokens. The summary back is ~600 words ≈ 800 output tokens.
$0.357 a month on GPT-5 Nano, $0.357 on GPT-5 Nano, and $69.00 on GPT-6 Astra. Batch APIs typically halve these figures for work that can wait, and prompt caching cuts the input side further when the same context is reused.
This is an input-heavy job, so the input price per million tokens decides almost the whole bill — the summary you get back is rounding error. Check the context window before the price: a model that can't hold 65,000 tokens has to be fed in chunks, which costs more and summarises worse.
Yes. Every figure on this page is computed from our model pricing data, which is checked daily against each provider's official pricing page. When a provider changes a price, these numbers change with it.
Newsletter
Pricing changes, new model releases, and updated recommendations — delivered when it matters.
No spam. Useful updates only. Affiliate disclosures always clearly labeled.