LLM prices / OpenAI / GPT-4.1
OpenAI API · prices checked Oct 23, 2026
GPT-4.1 API pricing
Short answerGPT-4.1 costs $2 per 1M input tokens and $8 per 1M output tokens on the OpenAI API (standard price, checked Oct 23, 2026). A typical chatbot reply (1,500 tokens in, 400 out) costs $0.0062, so 1,000 replies a day is about $186 a month and 1M replies is $6,200. OpenAI's own model page marks it deprecated, so it is left out of our picks.
Deprecated. OpenAI lists GPT-4.1 as deprecated on its model page. Prices stay here for old bills and migrations. The cheapest current OpenAI model for a chatbot reply is GPT-6 Luna.
Best for
- Bulk jobs that can wait: the batch API bills input at 50% of the standard price.
- Whole codebases or contracts in one prompt: 1.05M tokens of context.
- Reading more than text: takes image input.
Skip if
- You are starting something new: OpenAI marks it deprecated. The cheapest current OpenAI model for a chatbot reply is GPT-6 Luna.
Rules: Best for = cheapest quarter of current models for a chatbot reply, top 3 on one of our 8 jobs, cache reads at 15% or less of input, batch at 60% or less, 1M+ context, non-text inputs. Skip if = not current on the provider's page, priciest quarter, text only, no cache or batch price, a long-prompt surcharge, preview, under 200K context, no tool calling. Price and spec facts only; we don't score answer quality.
What your volume costs
Pick a job and a monthly volume. Same prices as everything on this page; caching and batch applied where they make it cheaper.
Cost per task, 8 jobs
Per 1,000 requests and per month at a typical volume, billed the cheapest way the price list allows. The rank is among current models that fit the job.
| Job (links to the cheapest model for it) | Per 1,000 | A month | Rank among fits |
|---|---|---|---|
| Chatbota 1,500-token prompt (system prompt plus the conversation so far) and a 400-token reply | $6.20standard price | $186at 30,000 a month | does not fit |
| RAG (answer from your docs)6,000 tokens of retrieved passages plus the question, and a 500-token answer | $16.00standard price | $480at 30,000 a month | does not fit |
| Coding agent40,000 tokens of repo context and history, 30,000 of them repeated from the last step and billed at the cache-read price, 2,000 tokens out, tool calling required | $51.00with the repeated 30,000 tokens billed at the cache-read price | $153at 3,000 a month | does not fit |
| Tool-using agent12,000 tokens of instructions, tool definitions and history, 10,000 of them repeated from the last step and cached, 600 tokens out, tool calling required | $13.80with the repeated 10,000 tokens billed at the cache-read price | $138at 10,000 a month | does not fit |
| Vision (read an image)one image counted as 1,000 input tokens plus a 300-token question, and a 300-token answer. Providers count image tokens differently, so treat vision costs as a rough guide | $5.00standard price | $50.00at 10,000 a month | does not fit |
| Translationa 1,000-token passage in and about 1,100 tokens out (translations run a little longer than the source) | $10.80standard price | $108at 10,000 a month | does not fit |
| Batch extraction2,000 tokens per record in and 300 tokens of JSON out, through the batch API where there is one (results within 24 hours) | $3.20through the batch API | $320at 100,000 a month | does not fit |
| Summarize long documentsa 20,000-token document in and an 800-token summary out | $46.40standard price | $46.40at 1,000 a month | does not fit |
Cheaper and pricier alternatives
The current models just below and just above GPT-4.1 on the cost of 1,000 chatbot replies, with what differs. Price only, not quality.
Cheaper
- DeepSeek V4 Pro $3.560.57x the cost of a chatbot reply, DeepSeek, 1M context vs 1.05M, text only.
- GPT-5.4 mini $2.930.47x the cost of a chatbot reply, same provider, 400K context vs 1.05M.
- Gemini 3.8 Flash $2.620.42x the cost of a chatbot reply, Google, 1.05M context vs 1.05M, takes image, video, audio, pdf too.
Pricier
- Claude Sonnet 5.5 $7.001.13x the cost of a chatbot reply, Anthropic, 1M context vs 1.05M.
- GPT-6.1 Sol $7.001.13x the cost of a chatbot reply, same provider, 1.05M context vs 1.05M.
- GPT-6 Sol $7.001.13x the cost of a chatbot reply, same provider, 1.05M context vs 1.05M.
Every price tier
| Input | $2 per 1M tokens |
| Output | $8 per 1M tokens |
| Cache read (hit) | $0.5 (25% of the input price) |
| Batch API | $1 input / $4 output |
| Fast tier | $3.5 input / $14 output |
| API model id | gpt-4.1 |
| Status on the pricing page | listed |
Provider notes: regional processing (data residency) adds 10% for models released on or after 2026-03-05
Model facts from OpenAI
| Context window | 1,047,576 tokens |
| Max output | 32,768 tokens |
| Input types | text, image |
| Tool calling | yes |
| Lifecycle | deprecated |
| Provider note | Page labels this model "Deprecated". |
Source: https://developers.openai.com/api/docs/models/gpt-4.1, checked Oct 24, 2026.
The 6 classic workloads
Per 1,000 requests at the standard price. "With caching" bills the repeated part of the prompt at the cache-read price (cache writes not counted); "via batch" uses the batch API price. Run your own token counts in the LLM API cost calculator.
| Workload | GPT-4.1per 1,000 requests |
|---|---|
| Chatbot replya 1,500-token prompt (1,000 of it a fixed system prompt) and a 400-token answer | $6.20$4.70 with caching |
| Answer from your docs (RAG)6,000 tokens of retrieved context plus the question, a 500-token answer | $16.00$14.50 with caching |
| Summarize a long documenta 20,000-token document in, an 800-token summary out | $46.40$44.90 with caching |
| Coding agent step40,000 tokens of repo context and history (30,000 of it repeated from the last step), 2,000 out | $96.00$51.00 with caching |
| Batch extraction2,000 tokens per record in, 300 tokens of JSON out, sent through the batch API where there is one | $6.40$3.20 via batch |
| Long-document Q&Aa 150,000-token contract or codebase in one prompt, a 1,000-token answer | $308 |
What the numbers say
- Output costs 4x input on GPT-4.1, so for answer-heavy jobs the output price drives the bill; for document-heavy jobs (RAG, summaries, agents) input does.
- Caching matters: a coding-agent step that rereads 30,000 cached tokens costs 47% less than the same step uncached ($51.00 vs $96.00 per 1,000 steps, cache writes not counted). How to stop paying twice for repeated prompts: LLM semantic cache.
- Among 41 listed OpenAI models it is the #17 cheapest for a chat reply.
- Same standard input/output price as o3; the differences are in caching, batch and long-prompt tiers.
Every listed OpenAI model, cheapest first
Same chatbot workload, 1,000 replies.
| Model | Input / 1M | Output / 1M | 1,000 chat replies |
|---|---|---|---|
| GPT-5 nanodeprecated | $0.05 | $0.4 | $0.23 |
| GPT-4.1 nanoshut down | $0.1 | $0.4 | $0.31 |
| GPT-6 Luna | $0.1 | $0.5 | $0.35 |
| GPT-4o minideprecated | $0.15 | $0.6 | $0.47 |
| GPT-5.6 Luna | $0.2 | $1.2 | $0.78 |
| GPT-5.4 nanodeprecated | $0.2 | $1.25 | $0.80 |
| GPT-5 minideprecated | $0.25 | $2 | $1.18 |
| GPT-4.1 minideprecated | $0.4 | $1.6 | $1.24 |
| GPT-3.5 Turboshut down | $0.5 | $1.5 | $1.35 |
| GPT-3.5 Turbo (0125)no spec page | $0.5 | $1.5 | $1.35 |
| GPT-5.4 mini | $0.75 | $4.5 | $2.93 |
| o4-minishut down | $1.1 | $4.4 | $3.41 |
| o3-minishut down | $1.1 | $4.4 | $3.41 |
| GPT-5.1deprecated | $1.25 | $10 | $5.88 |
| GPT-5deprecated | $1.25 | $10 | $5.88 |
| GPT-5 Search APIno spec page | $1.25 | $10 | $5.88 |
| GPT-4.1deprecated | $2 | $8 | $6.20 |
| o3deprecated | $2 | $8 | $6.20 |
| GPT-6.1 Sol | $2 | $10 | $7.00 |
| GPT-6 Sol | $2 | $10 | $7.00 |
| GPT-4odeprecated | $2.5 | $10 | $7.75 |
| GPT-5.6 Terra | $2 | $12 | $7.80 |
| GPT-5.2 | $1.75 | $14 | $8.22 |
| GPT-5.3 Codexdeprecated | $1.75 | $14 | $8.22 |
| GPT-5.4 | $2.5 | $15 | $9.75 |
| GPT-4o (2024-05-13)no spec page | $5 | $15 | $13.50 |
| GPT-5.6 Sol | $4 | $20 | $14.00 |
| GPT-5.5 | $5 | $30 | $19.50 |
| chat-latest | $5 | $30 | $19.50 |
| GPT-4 Turbo (2024-04-09)no spec page | $10 | $30 | $27.00 |
| GPT-6 Astra | $10 | $50 | $35.00 |
| o1shut down | $15 | $60 | $46.50 |
| GPT-5.6 Cyber | $12.5 | $75 | $48.75 |
| GPT-5.5 Cyberno spec page | $12.5 | $75 | $48.75 |
| o3-prodeprecated | $20 | $80 | $62.00 |
| GPT-4 (0613)no spec page | $30 | $60 | $69.00 |
| GPT-5 Prodeprecated | $15 | $120 | $70.50 |
| GPT-5.2 Pro | $21 | $168 | $98.70 |
| GPT-5.5 Pro | $30 | $180 | $117 |
| GPT-5.4 Pro | $30 | $180 | $117 |
| o1-proshut down | $150 | $600 | $465 |
Closest price from other providers
The current model from each other provider whose chatbot-reply cost is nearest. Price only: we do not rank quality here.
| Model | Input / 1M | Output / 1M | 1,000 chat replies | vs GPT-4.1 |
|---|---|---|---|---|
| Claude Sonnet 5.5Anthropic | $2 | $10 | $7.00 | 1.13x |
| Gemini 4 ArgonGoogle | $2 | $10 | $7.00 | 1.13x |
| DeepSeek V4 ProDeepSeek | $1.32 | $3.96 | $3.56 | 0.57x |
Price history
- Oct 23, 2026: GPT-4.1 added to the table: $2 input / $8 output per 1M tokens.
- Checked against the provider page on 1 date: Oct 23, 2026.
Every change we record is on the LLM price changelog.
Sources
- OpenAI API pricing: https://developers.openai.com/api/docs/pricing, checked Oct 23, 2026.
- Context, inputs and lifecycle: https://developers.openai.com/api/docs/models/gpt-4.1, checked Oct 24, 2026.
- Dataset: llm-prices (every row names its page and check date) and /llm/prices.csv.
Prices change often. If one is out of date, email matt@tensorust.com and we will fix it.