LLM API pricing statistics: cheapest and median
Published 2026-10-05 · Prose last reviewed 2026-11-08 · Figures recomputed nightly from the 2026-11-21 catalogue snapshot
A FinOps guide by OptimNow · prices from the live OptimToken catalogue
This page puts numbers on LLM API pricing: how many models carry a token price, which are the cheapest and the dearest, what the median model costs, and how wide each price tier is. Every figure is recomputed from the OptimToken catalogue when it refreshes, once a day. Each sentence that states one carries the date of the snapshot it describes, which is 2026-11-21 for this version.
These are statistics of the prices OptimToken quotes: one host's rate per model, in USD per 1M tokens, at list price. What that covers and what it leaves out is set out under How these figures are calculated.
Key figures on 2026-11-21
- As of 2026-11-21, the OptimToken catalogue quotes a token price for 306 LLM APIs. Leaving out 3 outlier listings, the 303 counted on this page come from 46 model makers: 32 Premium, 114 Mid-range and 157 Budget.
- As of 2026-11-21, the cheapest of 303 token-priced LLM APIs in the OptimToken catalogue, leaving out 3 outlier listings, is Mistral Nemo (Mistral), quoted at $0.029 per 1M input tokens and $0.03 per 1M output tokens, which is $0.03 on the blended price (30% input, 70% output).
- As of 2026-11-21, the most expensive of 303 token-priced LLM APIs in the OptimToken catalogue, leaving out 3 outlier listings, is GPT-5.4 Pro (OpenAI), level with 1 other model, quoted at $30.00 per 1M input tokens and $180.00 per 1M output tokens, which is $135.00 on the blended price (30% input, 70% output).
- As of 2026-11-21, the median price across 303 token-priced LLM APIs in the OptimToken catalogue, leaving out 3 outlier listings, is $0.44 per 1M input tokens and $1.80 per 1M output tokens, and the median blended price (30% input, 70% output) is $1.39.
- As of 2026-11-21, leaving out 3 outlier listings, quoted prices run from $0.015 to $30.00 per 1M input tokens and from $0.03 to $180.00 per 1M output tokens.
- As of 2026-11-21, the median blended price (30% input, 70% output) per 1M tokens is $0.49 for the 83 open-source models, $0.80 for the 41 open-weights models, $7.38 for the 111 proprietary models and $1.18 for the 68 models with no established licence.
- As of 2026-11-21, a batch rate is published for 81 of 303 token-priced models (27%) and a prompt-cache read rate for 220 (73%); 74 carry both and 76 carry neither.
How many models, and who makes them
| Price tier | Models | Share | Model makers |
|---|---|---|---|
| All tiers | 303 | 100% | 46 |
| Premium | 32 | 11% | 7 |
| Mid-range | 114 | 38% | 26 |
| Budget | 157 | 52% | 37 |
As of 2026-11-21, the model makers with the most token-priced models are OpenAI (50), Alibaba (40) and Google (25).
As of 2026-11-21, the prices OptimToken quotes for GLM 5.3 (Zhipu), Kimi K3.1 Pro (Moonshot) and Kimi K3 (Moonshot) are outlier listings. An outlier listing is one host's rate with input near free and output at or above the going rate, not what most hosts charge, so a cost built on it can be far from what you would pay. Such quoted prices are left out of these figures.
The catalogue also lists 8 image models. An image model's rate is not comparable with a text model's token price, so none enters the figures on this page. Also left out: 2 rows with no usable price on one side.
What a million tokens costs
As of 2026-11-21, the median price per 1M tokens by price tier is: Premium, $5.00 input and $27.50 output (32 models); Mid-range, $1.25 input and $4.75 output (114 models); Budget, $0.15 input and $0.50 output (157 models). As of 2026-11-21, the median token-priced model is quoted 4.0x as much for output tokens as for input tokens.
| Price tier | Models | Median input | Median output | Median blended | Input range | Output range |
|---|---|---|---|---|---|---|
| All tiers | 303 | $0.44 | $1.80 | $1.39 | $0.015 to $30.00 | $0.03 to $180.00 |
| Premium | 32 | $5.00 | $27.50 | $20.75 | $2.50 to $30.00 | $15.00 to $180.00 |
| Mid-range | 114 | $1.25 | $4.75 | $3.67 | $0.25 to $4.35 | $2.00 to $14.00 |
| Budget | 157 | $0.15 | $0.50 | $0.39 | $0.015 to $1.00 | $0.03 to $1.96 |
The blended price (30% input, 70% output) is the mix the pricing table ranks on. A tier is a price band, assigned from the quoted price alone. It is not a quality grade: the cheapest Premium model is the lowest price in the top band, whatever it scores. Which models beat their price band is the subject of the quality per dollar guide.
As of 2026-11-21, 130 of 303 token-priced models (43%) are quoted below $1 per 1M tokens on the blended price.
The cheapest and the dearest models
As of 2026-11-21, the 3 cheapest token-priced LLM APIs on the blended price (30% input, 70% output) are Mistral Nemo (Mistral) at $0.03, Llama 3.1 8B Instruct (Meta) at $0.034 and Ling 3.0 Tiny (inclusionai) at $0.036 per 1M tokens.
| Model | Model maker | Input, per 1M | Output, per 1M | Blended | Quoted host |
|---|---|---|---|---|---|
| Mistral Nemo | Mistral | $0.029 | $0.03 | $0.03 | DeepInfra (fp8 build) |
| Llama 3.1 8B Instruct | Meta | $0.02 | $0.04 | $0.034 | DeepInfra (fp8 build) |
| Ling 3.0 Tiny | inclusionai | $0.015 | $0.045 | $0.036 | Ant Digital (fp8 build) |
| Ling 3.0 Flash | inclusionai | $0.021 | $0.063 | $0.05 | Novita |
| gpt-oss-20b | OpenAI | $0.017 | $0.085 | $0.065 | Darkbloom (fp8 build) |
The price quoted for Mistral Nemo on 2026-11-21 is DeepInfra's rate for a quantized fp8 build, one of 5 hosts selling the model. Mistral, the model maker, charges $0.15 per 1M input tokens and $0.15 per 1M output tokens. Mistral Nemo is deprecated: its maker has announced its retirement, so it is a price to migrate off, not one to start on.
Each figure is the price OptimToken quotes for the model, which is one host's rate: another host can charge more or less for the same model. Each model's page lists every host selling it; the quoted hosts above are from the 2026-11-21 read. Where a model runs, and what the channel changes, is covered in a guide of its own.
As of 2026-11-21, the cheapest model in each price tier, ranked on the blended price and quoted per 1M tokens, is: Premium, GPT-5.4 (OpenAI) at $2.50 input and $15.00 output; Mid-range, Seed-2.0-Lite (ByteDance) at $0.25 input and $2.00 output, level with 2 other models; Budget, Mistral Nemo (Mistral) at $0.029 input and $0.03 output.
The cheapest model of each tier, on each of the 3 prices:
| Tier | On input, per 1M | On output, per 1M | On the blended price |
|---|---|---|---|
| All tiers | Ling 3.0 Tiny $0.015 | Mistral Nemo $0.03 | Mistral Nemo $0.03 |
| Premium | GPT-5.4 $2.50 | GPT-5.4 $15.00 (6 others at this price) | GPT-5.4 $11.25 |
| Mid-range | Seed-2.0-Lite $0.25 (2 others at this price) | Seed-2.0-Lite $2.00 (5 others at this price) | Seed-2.0-Lite $1.47 (2 others at this price) |
| Budget | Ling 3.0 Tiny $0.015 | Mistral Nemo $0.03 | Mistral Nemo $0.03 |
As of 2026-11-21, the lowest input price, $0.015 per 1M tokens, is quoted for Ling 3.0 Tiny (inclusionai), whose output is quoted at $0.045 per 1M tokens: 2 of the 303 models are cheaper on the blended price.
And the dearest:
| Tier | On input, per 1M | On output, per 1M | On the blended price |
|---|---|---|---|
| All tiers | GPT-5.4 Pro $30.00 (1 other at this price) | GPT-5.4 Pro $180.00 (1 other at this price) | GPT-5.4 Pro $135.00 (1 other at this price) |
| Premium | GPT-5.4 Pro $30.00 (1 other at this price) | GPT-5.4 Pro $180.00 (1 other at this price) | GPT-5.4 Pro $135.00 (1 other at this price) |
| Mid-range | MiMo-V2.6-Pro-UltraSpeed $4.35 | GPT-5.2 $14.00 (3 others at this price) | GPT-5.2 $10.32 (3 others at this price) |
| Budget | Sonar $1.00 | Qwen3.8 27B $1.96 | Morph V3 Large $1.60 |
Open source, open weights and proprietary
Openness is read from the licence, and it is a separate axis from the price tier. Open source means published weights under an OSI-approved licence with no field-of-use limit. Open weights means published weights under a licence that restricts something. Proprietary means API only. A model with no licence on record is counted on its own, as unknown, and assumed to be neither.
| Openness | Models | Share | Median input | Median output | Median blended |
|---|---|---|---|---|---|
| Open source | 83 | 27% | $0.15 | $0.60 | $0.49 |
| Open weights | 41 | 14% | $0.36 | $1.00 | $0.80 |
| Proprietary | 111 | 37% | $1.75 | $10.00 | $7.38 |
| Unknown | 68 | 22% | $0.35 | $1.50 | $1.18 |
As of 2026-11-21, the median open-weights model is quoted below the median proprietary model, which costs 9.2x as much on the blended price.
Each of those medians is taken on the price OptimToken quotes, which is one host's rate per model. On the 2026-11-21 read of host prices, among the models sold by more than one host, the dearest host charged a median 1.9x the cheapest across 51 open-source models, 1.4x the cheapest across 17 open-weights models, 1.00x the cheapest across 65 proprietary models and 1.6x the cheapest across 8 models with no established licence.
As of 2026-11-21, none of the 32 models in the Premium price tier is open source or open weights: 28 are proprietary and 4 carry no established licence.
These medians compare groups of different models, at whatever quality each group holds. The comparison at matched quality, and what a licence buys beyond the price, is in Open weights against proprietary APIs.
Batch and cache rates
As of 2026-11-21, a batch rate is published for 81 of 303 token-priced models (27%) and a prompt-cache read rate for 220 (73%); 74 carry both and 76 carry neither. A cache write rate, the price of storing a prompt in the cache before it can be read back, is published for 62 of them.
When each discount applies, and what it is worth on a real workload, is covered in Batch API economics and Prompt caching.
What moved
From 14 Nov to 21 Nov 2026: 7 price changes (4 up, 3 down) among 306 models compared. 24 other moves are flagged as a change of quoted host or an outlier listing.
Since tracking began on 2026-08-15, the price OptimToken quotes has changed at least once for 88 of the 386 models tracked: 765 changes in all, up to 2026-11-21. Of those changes, 390 are established as a change of quoted host, a test that needs both days' host prices, archived from 2026-08-30. For the other 298 models, the quoted price has not changed since the first day in the archive.
A price change, here as on the barometer, is a move in the price we quote that is not flagged as a change of quoted host or as an outlier listing. The price barometer lists every move of the latest window with its thresholds, and each completed week keeps a page of its own.
How these figures are calculated
- Source. Prices are list prices in USD per 1M tokens, read once a day from OpenRouter's public model list. Negotiated rates, committed-use discounts and provisioned capacity are not visible in them.
- One price per model. OptimToken quotes one price per model: the rate of one host, the one OpenRouter uses as its reference for the model. A model can be sold by more than one host at more than one rate, so these are statistics of quoted prices, not of every host's price. On the 2026-11-21 read of host prices, 141 of the 302 token-priced models with host data were sold by more than one host, and for 36 of those the dearest host charged at least twice the cheapest.
- Population. The statistics cover the 303 models in the Premium, Mid-range and Budget tiers that carry a price on both input and output. Image models are left out, and so is a model whose quoted price is an outlier listing: input near free and output at or above what most hosts charge for it.
- Median. The middle value once the models are ranked on a price. With an even number of models it is the mean of the 2 middle values. The input, output and blended medians are computed separately, so the blended median is not the blend of the other 2.
- Blended price. One figure per model, weighted 30% input, 70% output. Your own workload has its own token mix: the cost per request methodology prices 8 of them.
- Model makers. Counted from the maker named in the catalogue, ignoring capitalisation.
- Refresh. The page is rebuilt from the catalogue once a day. A figure quoted from it is a figure of its date.
How to cite this page
The dataset behind this page is published under Creative Commons Attribution 4.0. Quote a figure with its date, because the figures change:
OptimToken by OptimNow, LLM API pricing statistics, snapshot of 2026-11-21, https://optimtoken.optimnow.io/guides/llm-pricing-statistics
The same catalogue is available as JSON at /api/llm-models and through the MCP server.
The FinOps reasoning behind this guide comes from OptimNow's open-source practice library: github.com/OptimNow/cloud-finops-skills. If you want these list prices set against what your own contracts charge, OptimNow runs that review.
Questions this guide answers
What is the cheapest LLM API?
As of 2026-11-21, the cheapest of 303 token-priced LLM APIs in the OptimToken catalogue, leaving out 3 outlier listings, is Mistral Nemo (Mistral), quoted at $0.029 per 1M input tokens and $0.03 per 1M output tokens, which is $0.03 on the blended price (30% input, 70% output). The price quoted for Mistral Nemo on 2026-11-21 is DeepInfra's rate for a quantized fp8 build, one of 5 hosts selling the model. Mistral, the model maker, charges $0.15 per 1M input tokens and $0.15 per 1M output tokens. As of 2026-11-21, the cheapest model in each price tier, ranked on the blended price and quoted per 1M tokens, is: Premium, GPT-5.4 (OpenAI) at $2.50 input and $15.00 output; Mid-range, Seed-2.0-Lite (ByteDance) at $0.25 input and $2.00 output, level with 2 other models; Budget, Mistral Nemo (Mistral) at $0.029 input and $0.03 output. As of 2026-11-21, the prices OptimToken quotes for GLM 5.3 (Zhipu), Kimi K3.1 Pro (Moonshot) and Kimi K3 (Moonshot) are outlier listings. An outlier listing is one host's rate with input near free and output at or above the going rate, not what most hosts charge, so a cost built on it can be far from what you would pay. Such quoted prices are left out of these figures.
How much do LLM APIs cost per million tokens?
As of 2026-11-21, the median price across 303 token-priced LLM APIs in the OptimToken catalogue, leaving out 3 outlier listings, is $0.44 per 1M input tokens and $1.80 per 1M output tokens, and the median blended price (30% input, 70% output) is $1.39. As of 2026-11-21, leaving out 3 outlier listings, quoted prices run from $0.015 to $30.00 per 1M input tokens and from $0.03 to $180.00 per 1M output tokens. As of 2026-11-21, the median price per 1M tokens by price tier is: Premium, $5.00 input and $27.50 output (32 models); Mid-range, $1.25 input and $4.75 output (114 models); Budget, $0.15 input and $0.50 output (157 models). As of 2026-11-21, the prices OptimToken quotes for GLM 5.3 (Zhipu), Kimi K3.1 Pro (Moonshot) and Kimi K3 (Moonshot) are outlier listings. An outlier listing is one host's rate with input near free and output at or above the going rate, not what most hosts charge, so a cost built on it can be far from what you would pay. Such quoted prices are left out of these figures.
Are open-weight models cheaper than proprietary ones?
As of 2026-11-21, the median open-weights model is quoted below the median proprietary model, which costs 9.2x as much on the blended price. As of 2026-11-21, the median blended price (30% input, 70% output) per 1M tokens is $0.49 for the 83 open-source models, $0.80 for the 41 open-weights models, $7.38 for the 111 proprietary models and $1.18 for the 68 models with no established licence. Each of those medians is taken on the price OptimToken quotes, which is one host's rate per model. On the 2026-11-21 read of host prices, among the models sold by more than one host, the dearest host charged a median 1.9x the cheapest across 51 open-source models, 1.4x the cheapest across 17 open-weights models, 1.00x the cheapest across 65 proprietary models and 1.6x the cheapest across 8 models with no established licence.
What is the most expensive LLM API?
As of 2026-11-21, the most expensive of 303 token-priced LLM APIs in the OptimToken catalogue, leaving out 3 outlier listings, is GPT-5.4 Pro (OpenAI), level with 1 other model, quoted at $30.00 per 1M input tokens and $180.00 per 1M output tokens, which is $135.00 on the blended price (30% input, 70% output).
How many LLM APIs publish batch and prompt-cache prices?
As of 2026-11-21, a batch rate is published for 81 of 303 token-priced models (27%) and a prompt-cache read rate for 220 (73%); 74 carry both and 76 carry neither. A cache write rate, the price of storing a prompt in the cache before it can be read back, is published for 62 of them.
How often do LLM API prices change?
Since tracking began on 2026-08-15, the price OptimToken quotes has changed at least once for 88 of the 386 models tracked: 765 changes in all, up to 2026-11-21. Of those changes, 390 are established as a change of quoted host, a test that needs both days' host prices, archived from 2026-08-30. From 14 Nov to 21 Nov 2026: 7 price changes (4 up, 3 down) among 306 models compared. 24 other moves are flagged as a change of quoted host or an outlier listing.
Your real cost also depends on caching and batching, negotiated discounts, marketplace agreements, and the contracts you already have. OptimNow audits exactly that.
Get a FinOps reviewData snapshot: 2026-11-21 · Compare all models · Cloud compute pricing · All guides