Live · updated 8 hr ago 348 Models Tracked

The LLM Price Index

Every major large language model, normalized to dollars per million tokens. Scored on value, cheapness, and frontier capability. Independent, automated, refreshed daily.

Models 348
Providers 20+
Free Tier Models 27
Median Price / Mtok $0.628

What this is

The LLM Price Index is a live, independently maintained price comparison for every large language model offered through the OpenRouter catalog. Input and output pricing is normalized to dollars per million tokens, blended 3:1 to produce a single comparable figure, and scored 0–10 on three axes: value, cheapness, and frontier capability. The cheapest paid model right now is Gemma 3n 4B at $0.025 per million tokens (3:1 blended).

Qwen3 Next 80B A3B Instruct (free)

7.9/10
Qwen · qwen/qwen3-next-80b-a3b-instruct:free
Free Input /Mtok
Free Output /Mtok
Free Blended 3:1

Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces. It targets complex tasks…

10.0Value
10.0Cheap
3.0Frontier
Reasoning Free Open weights 262K ctx Tools

LFM2.5-1.2B-Thinking (free)

7.8/10
Liquid · liquid/lfm-2.5-1.2b-thinking:free
Free Input /Mtok
Free Output /Mtok
Free Blended 3:1

LFM2.5-1.2B-Thinking is a lightweight reasoning-focused model optimized for agentic tasks, data extraction, and RAG—while still running comfortably on edge devices. It supports…

10.0Value
10.0Cheap
2.5Frontier
Reasoning Free Open weights 33K ctx

Nemotron Nano 9B V2 (free)

7.8/10
NVIDIA · nvidia/nemotron-nano-9b-v2:free
Free Input /Mtok
Free Output /Mtok
Free Blended 3:1

NVIDIA-Nemotron-Nano-9B-v2 is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and…

10.0Value
10.0Cheap
2.5Frontier
Reasoning Free Open weights 128K ctx Tools

gpt-oss-120b (free)

7.8/10
OpenAI · openai/gpt-oss-120b:free
Free Input /Mtok
Free Output /Mtok
Free Blended 3:1

gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates…

10.0Value
10.0Cheap
2.5Frontier
Reasoning Free Open weights 131K ctx Tools

Trinity Mini

7.5/10
Arcee Ai · arcee-ai/trinity-mini
$0.045 Input /Mtok
$0.150 Output /Mtok
$0.071 Blended 3:1

Trinity Mini is a 26B-parameter (3B active) sparse mixture-of-experts language model featuring 128 experts with 8 active per token. Engineered for efficient…

10.0Value
8.3Cheap
3.5Frontier
Reasoning Open weights 131K ctx Tools #23 cheapest

Llama 3.2 3B Instruct (free)

7.5/10
Meta · meta-llama/llama-3.2-3b-instruct:free
Free Input /Mtok
Free Output /Mtok
Free Blended 3:1

Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and…

10.0Value
10.0Cheap
1.5Frontier
Reasoning Free Open weights 131K ctx

Hermes 3 405B Instruct (free)

7.5/10
Nous Research · nousresearch/hermes-3-llama-3.1-405b:free
Free Input /Mtok
Free Output /Mtok
Free Blended 3:1

Hermes 3 is a generalist language model with many improvements over Hermes 2, including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn…

10.0Value
10.0Cheap
1.5Frontier
Reasoning Free Open weights 131K ctx

Trinity Large Thinking

7.2/10
Arcee Ai · arcee-ai/trinity-large-thinking
$0.220 Input /Mtok
$0.850 Output /Mtok
$0.378 Blended 3:1

Trinity Large Thinking is a powerful open source reasoning model from the team at Arcee AI. It shows strong performance in PinchBench,…

8.1Value
6.9Cheap
6.5Frontier
Reasoning Open weights 262K ctx Tools

Nemotron Nano 9B V2

7.2/10
NVIDIA · nvidia/nemotron-nano-9b-v2
$0.040 Input /Mtok
$0.160 Output /Mtok
$0.070 Blended 3:1

NVIDIA-Nemotron-Nano-9B-v2 is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and…

10.0Value
8.3Cheap
2.5Frontier
Reasoning Open weights 131K ctx Tools #21 cheapest

gpt-oss-120b

7.1/10
OpenAI · openai/gpt-oss-120b
$0.039 Input /Mtok
$0.190 Output /Mtok
$0.077 Blended 3:1

gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates…

10.0Value
8.2Cheap
2.5Frontier
Reasoning Open weights 131K ctx Tools #25 cheapest

Qwen Plus 0728 (thinking)

7.1/10
Qwen · qwen/qwen-plus-2025-07-28:thinking
$0.260 Input /Mtok
$0.780 Output /Mtok
$0.390 Blended 3:1

Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed,…

10.0Value
6.8Cheap
4.0Frontier
Reasoning 1.0M ctx Tools

Qwen Plus 0728

7.1/10
Qwen · qwen/qwen-plus-2025-07-28
$0.260 Input /Mtok
$0.780 Output /Mtok
$0.390 Blended 3:1

Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed,…

10.0Value
6.8Cheap
4.0Frontier
Reasoning 1.0M ctx Tools

Qwen3 30B A3B Instruct 2507

7.1/10
Qwen · qwen/qwen3-30b-a3b-instruct-2507
$0.090 Input /Mtok
$0.300 Output /Mtok
$0.143 Blended 3:1

Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference. It operates in non-thinking mode and is…

10.0Value
7.7Cheap
3.0Frontier
Reasoning Open weights 262K ctx Tools

gpt-oss-safeguard-20b

6.7/10
OpenAI · openai/gpt-oss-safeguard-20b
$0.075 Input /Mtok
$0.300 Output /Mtok
$0.131 Blended 3:1

gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b. This open-weight, 21B-parameter Mixture-of-Experts (MoE) model offers lower latency for safety…

8.2Value
7.8Cheap
3.5Frontier
Reasoning Open weights 131K ctx Tools #48 cheapest

Command R7B (12-2024)

6.6/10
Cohere · cohere/command-r7b-12-2024
$0.038 Input /Mtok
$0.150 Output /Mtok
$0.066 Blended 3:1

Command R7B (12-2024) is a small, fast update of the Command R+ model, delivered in December 2024. It excels at RAG, tool…

9.3Value
8.4Cheap
1.5Frontier
Reasoning 128K ctx #20 cheapest

Qwen3 30B A3B Thinking 2507

6.6/10
Qwen · qwen/qwen3-30b-a3b-thinking-2507
$0.080 Input /Mtok
$0.400 Output /Mtok
$0.160 Blended 3:1

Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking. The model is designed specifically for…

9.0Value
7.6Cheap
2.5Frontier
Reasoning Open weights 131K ctx Tools

Mercury 2

6.3/10
Inception · inception/mercury-2
$0.250 Input /Mtok
$0.750 Output /Mtok
$0.375 Blended 3:1

Mercury 2 is an extremely fast reasoning LLM, and the first reasoning diffusion LLM (dLLM). Instead of generating tokens sequentially, Mercury 2…

6.9Value
6.9Cheap
5.0Frontier
Reasoning 128K ctx Tools

Qwen3 14B

6.3/10
Qwen · qwen/qwen3-14b
$0.060 Input /Mtok
$0.240 Output /Mtok
$0.105 Blended 3:1

Qwen3-14B is a dense 14.8B parameter causal language model from the Qwen3 series, designed for both complex reasoning and efficient dialogue. It…

8.6Value
8.0Cheap
1.5Frontier
Reasoning Open weights 41K ctx Tools #34 cheapest

Qwen3 Next 80B A3B Instruct

6.2/10
Qwen · qwen/qwen3-next-80b-a3b-instruct
$0.090 Input /Mtok
$1.10 Output /Mtok
$0.343 Blended 3:1

Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces. It targets complex tasks…

8.3Value
6.9Cheap
3.0Frontier
Reasoning Open weights 262K ctx Tools

MiniMax M1

6.1/10
Minimax · minimax/minimax-m1
$0.400 Input /Mtok
$2.20 Output /Mtok
$0.850 Blended 3:1

MiniMax-M1 is a large-scale, open-weight reasoning model designed for extended context and high-efficiency inference. It leverages a hybrid Mixture-of-Experts (MoE) architecture paired…

7.9Value
6.1Cheap
4.0Frontier
Reasoning 1.0M ctx Tools

DeepSeek V3.2

6.1/10
DeepSeek · deepseek/deepseek-v3.2
$0.260 Input /Mtok
$0.380 Output /Mtok
$0.290 Blended 3:1

DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance. It introduces DeepSeek…

7.4Value
7.1Cheap
3.5Frontier
Reasoning Open weights 164K ctx Tools

Qwen3 32B

6.1/10
Qwen · qwen/qwen3-32b
$0.080 Input /Mtok
$0.240 Output /Mtok
$0.120 Blended 3:1

Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue. It…

8.2Value
7.8Cheap
1.5Frontier
Reasoning Open weights 41K ctx Tools #39 cheapest

Rnj 1 Instruct

6.1/10
Essentialai · essentialai/rnj-1-instruct
$0.150 Input /Mtok
$0.150 Output /Mtok
$0.150 Blended 3:1

Rnj-1 is an 8B-parameter, dense, open-weight model family developed by Essential AI and trained from scratch with a focus on programming, math,…

7.5Value
7.7Cheap
2.5Frontier
Reasoning Open weights 33K ctx Tools

Llama 3.3 Nemotron Super 49B V1.5

6.0/10
NVIDIA · nvidia/llama-3.3-nemotron-super-49b-v1.5
$0.100 Input /Mtok
$0.400 Output /Mtok
$0.175 Blended 3:1

Llama-3.3-Nemotron-Super-49B-v1.5 is a 49B-parameter, English-centric reasoning/chat model derived from Meta’s Llama-3.3-70B-Instruct with a 128K context. It’s post-trained for agentic workflows (RAG, tool…

7.4Value
7.5Cheap
2.5Frontier
Reasoning Open weights 131K ctx Tools

Qwen3 30B A3B

5.9/10
Qwen · qwen/qwen3-30b-a3b
$0.080 Input /Mtok
$0.280 Output /Mtok
$0.130 Blended 3:1

Qwen3, the latest generation in the Qwen large language model series, features both dense and mixture-of-experts (MoE) architectures to excel in reasoning,…

7.9Value
7.8Cheap
1.5Frontier
Reasoning Open weights 41K ctx Tools #45 cheapest

Qwen3 Next 80B A3B Thinking

5.9/10
Qwen · qwen/qwen3-next-80b-a3b-thinking
$0.098 Input /Mtok
$0.780 Output /Mtok
$0.268 Blended 3:1

Qwen3-Next-80B-A3B-Thinking is a reasoning-first chat model in the Qwen3-Next line that outputs structured “thinking” traces by default. It’s designed for hard multi-step…

7.6Value
7.1Cheap
2.5Frontier
Reasoning Open weights 131K ctx Tools

Qwen3 8B

5.9/10
Qwen · qwen/qwen3-8b
$0.050 Input /Mtok
$0.400 Output /Mtok
$0.138 Blended 3:1

Qwen3-8B is a dense 8.2B parameter causal language model from the Qwen3 series, designed for both reasoning-heavy tasks and efficient dialogue. It…

7.7Value
7.7Cheap
1.5Frontier
Reasoning Open weights 41K ctx Tools

GLM 4.7

5.8/10
Z Ai · z-ai/glm-4.7
$0.390 Input /Mtok
$1.75 Output /Mtok
$0.730 Blended 3:1

GLM-4.7 is Z.ai’s latest flagship model, featuring upgrades in two key areas: enhanced programming capabilities and more stable multi-step reasoning/execution. It demonstrates…

6.9Value
6.3Cheap
4.0Frontier
Reasoning Open weights 203K ctx Tools

Kimi K2 Thinking

5.8/10
Moonshotai · moonshotai/kimi-k2-thinking
$0.600 Input /Mtok
$2.50 Output /Mtok
$1.08 Blended 3:1

Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning. Built…

6.3Value
5.9Cheap
5.0Frontier
Reasoning Open weights 262K ctx Tools

MiniMax M2

5.7/10
Minimax · minimax/minimax-m2
$0.255 Input /Mtok
$1.00 Output /Mtok
$0.441 Blended 3:1

MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows. With 10 billion activated parameters (230 billion…

6.6Value
6.7Cheap
3.5Frontier
Reasoning Open weights 197K ctx Tools

Command R (08-2024)

5.6/10
Cohere · cohere/command-r-08-2024
$0.150 Input /Mtok
$0.600 Output /Mtok
$0.263 Blended 3:1

command-r-08-2024 is an update of the [Command R](/models/cohere/command-r) with improved performance for multilingual retrieval-augmented generation (RAG) and tool use. More broadly, it…

7.6Value
7.2Cheap
1.5Frontier
Reasoning 128K ctx Tools

Grok 3 Mini

5.6/10
xAI · x-ai/grok-3-mini
$0.300 Input /Mtok
$0.500 Output /Mtok
$0.350 Blended 3:1

A lightweight model that thinks before responding. Fast, smart, and great for logic-based tasks that do not require deep domain knowledge. The…

7.0Value
6.9Cheap
2.5Frontier
Reasoning 131K ctx Tools

Olmo 2 32B Instruct

5.5/10
Allenai · allenai/olmo-2-0325-32b-instruct
$0.050 Input /Mtok
$0.200 Output /Mtok
$0.088 Blended 3:1

OLMo-2 32B Instruct is a supervised instruction-finetuned variant of the OLMo-2 32B March 2025 base model. It excels in complex reasoning and…

6.4Value
8.1Cheap
1.5Frontier
Reasoning Open weights 128K ctx #29 cheapest

Qwen3 Max Thinking

5.5/10
Qwen · qwen/qwen3-max-thinking
$0.780 Input /Mtok
$3.90 Output /Mtok
$1.56 Blended 3:1

Qwen3-Max-Thinking is the flagship reasoning model in the Qwen3 series, designed for high-stakes cognitive tasks that require deep, multi-step reasoning. By significantly…

5.8Value
5.6Cheap
5.0Frontier
Reasoning 262K ctx Tools

ERNIE 4.5 21B A3B Thinking

5.4/10
Baidu · baidu/ernie-4.5-21b-a3b-thinking
$0.070 Input /Mtok
$0.280 Output /Mtok
$0.123 Blended 3:1

ERNIE-4.5-21B-A3B-Thinking is Baidu's upgraded lightweight MoE model, refined to boost reasoning depth and quality for top-tier performance in logical puzzles, math, science,…

5.5Value
7.8Cheap
2.5Frontier
Reasoning Open weights 131K ctx #40 cheapest

Qwen3 235B A22B Thinking 2507

5.3/10
Qwen · qwen/qwen3-235b-a22b-thinking-2507
$0.150 Input /Mtok
$1.50 Output /Mtok
$0.486 Blended 3:1

Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks. It activates 22B of its 235B parameters per…

6.4Value
6.6Cheap
2.5Frontier
Reasoning Open weights 131K ctx Tools

DeepSeek R1T2 Chimera

5.3/10
Tngtech · tngtech/deepseek-r1t2-chimera
$0.300 Input /Mtok
$1.10 Output /Mtok
$0.500 Blended 3:1

DeepSeek-TNG-R1T2-Chimera is the second-generation Chimera model from TNG Tech. It is a 671 B-parameter mixture-of-experts text-generation model assembled from DeepSeek-AI’s R1-0528, R1,…

6.4Value
6.6Cheap
2.5Frontier
Reasoning Open weights 164K ctx Tools

QwQ 32B

5.2/10
Qwen · qwen/qwq-32b
$0.150 Input /Mtok
$0.580 Output /Mtok
$0.258 Blended 3:1

QwQ is the reasoning model of the Qwen series. Compared with conventional instruction-tuned models, QwQ, which is capable of thinking and reasoning,…

6.5Value
7.2Cheap
1.5Frontier
Reasoning Open weights 131K ctx Tools

Hunyuan A13B Instruct

5.2/10
Tencent · tencent/hunyuan-a13b-instruct
$0.140 Input /Mtok
$0.570 Output /Mtok
$0.248 Blended 3:1

Hunyuan-A13B is a 13B active parameter Mixture-of-Experts (MoE) language model developed by Tencent, with a total parameter count of 80B and support…

5.5Value
7.2Cheap
2.5Frontier
Reasoning Open weights 131K ctx

Llama 3 8B Lunaris

5.1/10
Sao10K · sao10k/l3-lunaris-8b
$0.040 Input /Mtok
$0.050 Output /Mtok
$0.043 Blended 3:1

Lunaris 8B is a versatile generalist and roleplaying model based on Llama 3. It's a strategic merge of multiple models, designed to…

5.7Value
8.7Cheap
0.0Frontier
Reasoning Open weights 8K ctx #7 cheapest

DeepSeek V3.1

5.0/10
DeepSeek · deepseek/deepseek-chat-v3.1
$0.150 Input /Mtok
$0.750 Output /Mtok
$0.300 Blended 3:1

DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates. It…

6.0Value
7.1Cheap
1.5Frontier
Reasoning Open weights 33K ctx Tools

Hermes 4 70B

5.0/10
Nous Research · nousresearch/hermes-4-70b
$0.130 Input /Mtok
$0.400 Output /Mtok
$0.198 Blended 3:1

Hermes 4 70B is a hybrid reasoning model from Nous Research, built on Meta-Llama-3.1-70B. It introduces the same hybrid mode as the…

4.6Value
7.4Cheap
2.5Frontier
Reasoning Open weights 131K ctx

Grok 3 Mini Beta

5.0/10
xAI · x-ai/grok-3-mini-beta
$0.300 Input /Mtok
$0.500 Output /Mtok
$0.350 Blended 3:1

Grok 3 Mini is a lightweight, smaller thinking model. Unlike traditional models that generate answers immediately, Grok 3 Mini thinks before responding.…

6.0Value
6.9Cheap
1.5Frontier
Reasoning 131K ctx Tools

Qwen3 Max

5.0/10
Qwen · qwen/qwen3-max
$0.780 Input /Mtok
$3.90 Output /Mtok
$1.56 Blended 3:1

Qwen3-Max is an updated release built on the Qwen3 series, offering major improvements in reasoning, instruction following, multilingual support, and long-tail knowledge…

5.1Value
5.6Cheap
4.0Frontier
Reasoning 262K ctx Tools

Devstral Medium

4.9/10
Mistral · mistralai/devstral-medium
$0.400 Input /Mtok
$2.00 Output /Mtok
$0.800 Blended 3:1

Devstral Medium is a high-performance code generation and agentic reasoning model developed jointly by Mistral AI and All Hands AI. Positioned as…

5.7Value
6.2Cheap
2.5Frontier
Reasoning 131K ctx Tools

R1 0528

4.8/10
DeepSeek · deepseek/deepseek-r1-0528
$0.450 Input /Mtok
$2.15 Output /Mtok
$0.875 Blended 3:1

May 28th update to the [original DeepSeek R1](/deepseek/deepseek-r1) Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens.…

5.6Value
6.1Cheap
2.5Frontier
Reasoning Open weights 164K ctx Tools

DeepSeek V3.2 Speciale

4.8/10
DeepSeek · deepseek/deepseek-v3.2-speciale
$0.400 Input /Mtok
$1.20 Output /Mtok
$0.600 Blended 3:1

DeepSeek-V3.2-Speciale is a high-compute variant of DeepSeek-V3.2 optimized for maximum reasoning and agentic performance. It builds on DeepSeek Sparse Attention (DSA) for…

4.3Value
6.4Cheap
3.5Frontier
Reasoning Open weights 164K ctx

o3 Mini High

4.8/10
OpenAI · openai/o3-mini-high
$1.10 Input /Mtok
$4.40 Output /Mtok
$1.93 Blended 3:1

OpenAI o3-mini-high is the same model as [o3-mini](/openai/o3-mini) with reasoning_effort set to high. o3-mini is a cost-efficient language model optimized for STEM…

5.6Value
5.4Cheap
3.0Frontier
Reasoning 200K ctx Tools

Free · No spam

Price drops, new models, deprecations

Every Tuesday: which models got cheaper, which launched, which got pulled. Five minutes. No filler.

Three axes, one overall score

  • Value (35%) — capability per dollar. Context length, vision, tools, and structured-output support divided by log-scaled blended price.
  • Cheapness (35%) — raw affordability. Free models score 10. Paid models use an inverse log curve anchored at $0.01 / Mtok.
  • Frontier (30%) — how close to the state of the art. Recent releases, long context windows, and premium pricing all contribute.

Blended price formula

  • Most production workloads are input-heavy, so the index uses a 3:1 blended price: (input × 0.75) + (output × 0.25).
  • All prices are normalized to dollars per million tokens. OpenRouter publishes per-token figures which we multiply by 1,000,000 before display.

Where does the pricing data come from?

Every model and price on this page is sourced from OpenRouter's public models API, which aggregates pricing directly from model providers including Anthropic, OpenAI, Google, Mistral, Meta, xAI, DeepSeek, and dozens of others. The pipeline re-fetches and re-scores the entire catalog once per day.

Why normalize to $/million tokens?

Model providers publish prices in inconsistent units — per 1K tokens, per million tokens, per character, sometimes per request. Comparing them directly is error-prone. Dollars per million tokens is the industry's most common reporting unit and makes cross-provider comparisons immediate and honest.

What does "3:1 blended" mean?

Most production LLM workloads are input-heavy — context, RAG retrievals, system prompts — while output is comparatively short. A 3:1 input:output ratio is the informal industry convention for producing a single number that reflects typical cost: (input × 0.75) + (output × 0.25). Your actual ratio may differ; always check both input and output columns for workloads with long generations.

What's the cheapest LLM right now?

The cheapest paid model as of the latest scan is Gemma 3n 4B from Google at $0.025 per million tokens (3:1 blended). Sort by "Cheapest" above for the full ranking. Many providers also offer free-tier variants of their models, which score a perfect 10 on the cheapness axis.

Is this affiliated with OpenRouter or any provider?

No. MegaOne AI is independent. OpenRouter is used as a public data source because their models API is the most complete and up-to-date LLM catalog available, but this directory is not operated by OpenRouter and we rate all models — including ones that compete with one another.

How often does the price index update?

A full re-fetch, re-score, and daily snapshot runs once per 24 hours. Snapshots are written to a history table so we can build price-over-time charts and detect drops. New models typically appear within 24 hours of being added to OpenRouter.

Is it free to use?

Yes. Browsing, filtering, sorting, and searching the entire price index is free. The weekly email briefing is free. There is no account required and no paywall.