AI pricing index / Model usage

Understand the rate.
Then model your workload.

Compare input and output token prices across every major provider. These are model rates, separate from Swfte product subscriptions and platform fees.

56

Models Tracked

17

Providers

$0.10

Cheapest Input

300x

Price Range

Estimate a workload

56 of 56 models

Model rates as recorded on Sep 15, 2026. Review dates and assumptions before making a decision. Swfte product subscriptions and Connect platform fees are priced separately.

Explore
Gemma 4 27B OSSGoogleSelf-hostSelf-hostSelf-host750.0128KProfile
Nemotron 3 Nano Omni OSSNVIDIASelf-hostSelf-hostSelf-host760.0256KProfile
Qwen3.8 27B OSSAlibaba CloudSelf-hostSelf-hostSelf-host800.0262KProfile
Nemotron 3.5 Lightning OSSNVIDIASelf-hostSelf-hostSelf-host790.01MProfile
Hunyuan Hy3 OSSTencentSelf-hostSelf-hostSelf-host760.0256KProfile
Ling-3.0-Flash OSSAnt GroupSelf-hostSelf-hostSelf-host740.0256KProfile
DeepSeek V4 Flash OSSDeepSeek$0.14$0.28$0.2184400.01MProfile
Gemini 2.0 FlashGoogle$0.10$0.40$0.2574296.01MProfile
Llama 4 Scout OSSMeta$0.15$0.40$0.2871258.210MProfile
Qwen 2.5 Coder 32B OSSAlibaba Cloud$0.15$0.45$0.3074246.7131KProfile
GPT-4o MiniOpenAI$0.15$0.60$0.3872192.0128KProfile
Llama 4 Maverick OSSMeta$0.20$0.60$0.4080200.01MProfile
Grok 3 MinixAI$0.30$0.50$0.4078195.0131KProfile
CodestralMistral AI$0.30$0.90$0.6076126.7256KProfile
Qwen 2.5 72B OSSAlibaba Cloud$0.30$0.90$0.6080133.3131KProfile
DeepSeek V3 OSSDeepSeek$0.27$1.10$0.6986125.5128KProfile
GPT-5.6 Luna 80%OpenAI$0.20$1.20$0.7083118.61MProfile
DeepSeek V4 Pro OSSDeepSeek$0.66$1.98$1.328967.41MProfile
DeepSeek R1 OSSDeepSeek$0.55$2.19$1.379166.4128KProfile
Gemini 3.5 Flash-LiteGoogle$0.30$2.50$1.407956.41MProfile
MiniMax M3 OSSMiniMax$0.60$2.40$1.508959.31MProfile
Grok 4.3xAI$1.25$2.50$1.889349.61MProfile
Amazon Nova ProAmazon$0.80$3.20$2.007035.0300KProfile
GLM-5.1 OSSZ.ai (Zhipu AI)$0.98$3.08$2.038843.3200KProfile
Kimi K2.6Moonshot AI$0.73$3.49$2.119243.6256KProfile
Gemini 3.6 FlashGoogle$0.75$3.75$2.259040.01MProfile
Claude 3.5 Haiku 20%Anthropic$0.80$4.00$2.407531.3200KProfile
o3 MiniOpenAI$1.10$4.40$2.758832.0200KProfile
GLM-5.2 OSSZ.ai (Zhipu AI)$1.40$4.40$2.909131.41MProfile
Qwen 3.6 PlusAlibaba Cloud$1.40$5.60$3.508624.6256KProfile
Mistral Large 2 33%Mistral AI$2.00$6.00$4.007919.8128KProfile
Grok 4.5xAI$2.00$6.00$4.009423.5500KProfile
Qwen3.8 Max OSSAlibaba Cloud$2.00$6.00$4.009624.01MProfile
o3 80%OpenAI$2.00$8.00$5.009418.8200KProfile
GPT-4.1OpenAI$2.00$8.00$5.008917.81MProfile
Qwen 3.7 MaxAlibaba Cloud$2.50$7.50$5.009418.81MProfile
Gemini 2.5 ProGoogle$1.25$10.00$5.639216.41MProfile
GPT-4oOpenAI$2.50$10.00$6.258513.6128KProfile
Command R+Cohere$2.50$10.00$6.256810.9128KProfile
Gemini 3.1 ProGoogle$2.00$12.00$7.009613.71MProfile
GPT-5.6 Terra 20%OpenAI$2.00$12.00$7.009313.31MProfile
Claude Sonnet 4Anthropic$3.00$15.00$9.00889.8200KProfile
Grok 3xAI$3.00$15.00$9.00879.7131KProfile
Sonar ProPerplexity$3.00$15.00$9.00788.7200KProfile
Kimi K3 OSSMoonshot AI$3.00$15.00$9.009710.81MProfile
Claude Sonnet 4.6Anthropic$3.00$15.00$9.009010.01MProfile
Claude Sonnet 5Anthropic$3.00$15.00$9.009310.31MProfile
Claude Opus 4.7Anthropic$5.00$25.00$15.00966.41MProfile
Claude Opus 4.8Anthropic$5.00$25.00$15.00986.51MProfile
Claude Opus 5Anthropic$5.00$25.00$15.00996.61MProfile
GPT-5.5OpenAI$5.00$30.00$17.50975.51MProfile
GPT-5.6 SolOpenAI$5.00$30.00$17.50985.61MProfile
Claude Fable 5Anthropic$10.00$50.00$30.001003.31MProfile
Claude Mythos 5Anthropic$10.00$50.00$30.001003.31MProfile
Claude Opus 4Anthropic$15.00$75.00$45.00912.0200KProfile
GPT-5.5 ProOpenAI$30.00$180.00$105.00960.91MProfile

Blended is the average of the input and output rate per 1M tokens. Quality is a composite benchmark score out of 100. Value is quality per dollar — higher is better. Self-hosted models carry no per-token vendor rate.

Estimate Your Monthly Cost

Monthly cost estimate

Enter your typical request shape. Costs below are projected over one month, based on current public list-price API rates.

Per month: 100K requests · 50.0M input tokens · 30.0M output tokens. Excludes prompt caching, batch discounts, retries, and fees.

Cheapest

DeepSeek V4 Flash

$15.40

per month at this volume

Best value (quality ≥ 80)

DeepSeek V4 Flash · Q 84

$15.40

per month at this volume

Most expensive

GPT-5.5 Pro

$6900.00

per month at this volume

Save 30-60% with Mixture-of-Routers

Most production traffic is mixed-difficulty. Send the easy 60% to a cheap model and the hard 10% to a frontier model: same quality, fraction of the cost.

See the math

Full breakdown by model

Sorted cheapest to most expensive

ModelCost / requestInput cost / moOutput cost / moTotal / mo

DeepSeek V4 Flash

$0.14 in / $0.28 out per 1M

$0.000154$7.00$8.40$15.40

Gemini 2.0 Flash

$0.1 in / $0.4 out per 1M

$0.000170$5.00$12.00$17.00

Llama 4 Scout

$0.15 in / $0.4 out per 1M

$0.000195$7.50$12.00$19.50

Qwen 2.5 Coder 32B

$0.15 in / $0.45 out per 1M

$0.000210$7.50$13.50$21.00

GPT-4o Mini

$0.15 in / $0.6 out per 1M

$0.000255$7.50$18.00$25.50

Llama 4 Maverick

$0.2 in / $0.6 out per 1M

$0.000280$10.00$18.00$28.00

Grok 3 Mini

$0.3 in / $0.5 out per 1M

$0.000300$15.00$15.00$30.00

Codestral

$0.3 in / $0.9 out per 1M

$0.000420$15.00$27.00$42.00

Qwen 2.5 72B

$0.3 in / $0.9 out per 1M

$0.000420$15.00$27.00$42.00

GPT-5.6 Luna

$0.2 in / $1.2 out per 1M

$0.000460$10.00$36.00$46.00

DeepSeek V3

$0.27 in / $1.1 out per 1M

$0.000465$13.50$33.00$46.50

Gemini 3.5 Flash-Lite

$0.3 in / $2.5 out per 1M

$0.000900$15.00$75.00$90.00

DeepSeek V4 Pro

$0.66 in / $1.98 out per 1M

$0.000924$33.00$59.40$92.40

DeepSeek R1

$0.55 in / $2.19 out per 1M

$0.000932$27.50$65.70$93.20

MiniMax M3

$0.6 in / $2.4 out per 1M

$0.001020$30.00$72.00$102.00

Amazon Nova Pro

$0.8 in / $3.2 out per 1M

$0.001360$40.00$96.00$136.00

Grok 4.3

$1.25 in / $2.5 out per 1M

$0.001375$62.50$75.00$137.50

Kimi K2.6

$0.73 in / $3.49 out per 1M

$0.001412$36.50$104.70$141.20

GLM-5.1

$0.98 in / $3.08 out per 1M

$0.001414$49.00$92.40$141.40

Gemini 3.6 Flash

$0.75 in / $3.75 out per 1M

$0.001500$37.50$112.50$150.00

Claude 3.5 Haiku

$0.8 in / $4 out per 1M

$0.001600$40.00$120.00$160.00

o3 Mini

$1.1 in / $4.4 out per 1M

$0.001870$55.00$132.00$187.00

GLM-5.2

$1.4 in / $4.4 out per 1M

$0.002020$70.00$132.00$202.00

Qwen 3.6 Plus

$1.4 in / $5.6 out per 1M

$0.002380$70.00$168.00$238.00

Mistral Large 2

$2 in / $6 out per 1M

$0.002800$100.00$180.00$280.00

Grok 4.5

$2 in / $6 out per 1M

$0.002800$100.00$180.00$280.00

Qwen3.8 Max

$2 in / $6 out per 1M

$0.002800$100.00$180.00$280.00

o3

$2 in / $8 out per 1M

$0.003400$100.00$240.00$340.00

GPT-4.1

$2 in / $8 out per 1M

$0.003400$100.00$240.00$340.00

Qwen 3.7 Max

$2.5 in / $7.5 out per 1M

$0.003500$125.00$225.00$350.00

Gemini 2.5 Pro

$1.25 in / $10 out per 1M

$0.003625$62.50$300.00$362.50

GPT-4o

$2.5 in / $10 out per 1M

$0.004250$125.00$300.00$425.00

Command R+

$2.5 in / $10 out per 1M

$0.004250$125.00$300.00$425.00

Gemini 3.1 Pro

$2 in / $12 out per 1M

$0.004600$100.00$360.00$460.00

GPT-5.6 Terra

$2 in / $12 out per 1M

$0.004600$100.00$360.00$460.00

Claude Sonnet 4

$3 in / $15 out per 1M

$0.006000$150.00$450.00$600.00

Grok 3

$3 in / $15 out per 1M

$0.006000$150.00$450.00$600.00

Sonar Pro

$3 in / $15 out per 1M

$0.006000$150.00$450.00$600.00

Kimi K3

$3 in / $15 out per 1M

$0.006000$150.00$450.00$600.00

Claude Sonnet 4.6

$3 in / $15 out per 1M

$0.006000$150.00$450.00$600.00

Claude Sonnet 5

$3 in / $15 out per 1M

$0.006000$150.00$450.00$600.00

Claude Opus 4.7

$5 in / $25 out per 1M

$0.0100$250.00$750.00$1000.00

Claude Opus 4.8

$5 in / $25 out per 1M

$0.0100$250.00$750.00$1000.00

Claude Opus 5

$5 in / $25 out per 1M

$0.0100$250.00$750.00$1000.00

GPT-5.5

$5 in / $30 out per 1M

$0.0115$250.00$900.00$1150.00

GPT-5.6 Sol

$5 in / $30 out per 1M

$0.0115$250.00$900.00$1150.00

Claude Fable 5

$10 in / $50 out per 1M

$0.0200$500.00$1500.00$2000.00

Claude Mythos 5

$10 in / $50 out per 1M

$0.0200$500.00$1500.00$2000.00

Claude Opus 4

$15 in / $75 out per 1M

$0.0300$750.00$2250.00$3000.00

GPT-5.5 Pro

$30 in / $180 out per 1M

$0.0690$1500.00$5400.00$6900.00

Gemma 4 27B

Self-host

Open weights (Apache 2.0) , token cost is $0; infra cost depends on hardware

Self-host

Nemotron 3 Nano Omni

Self-host

Open weights (NVIDIA Open Model License) , token cost is $0; infra cost depends on hardware

Self-host

Qwen3.8 27B

Self-host

Open weights (Apache 2.0) , token cost is $0; infra cost depends on hardware

Self-host

Nemotron 3.5 Lightning

Self-host

Open weights (OpenMDW-1.1) , token cost is $0; infra cost depends on hardware

Self-host

Hunyuan Hy3

Self-host

Open weights (Apache 2.0) , token cost is $0; infra cost depends on hardware

Self-host

Ling-3.0-Flash

Self-host

Open weights (MIT) , token cost is $0; infra cost depends on hardware

Self-host

List-price estimate. Real bills typically run 1.3-1.7x higher after retries, system-prompt re-sends, and tool-call round-trips. See per-million-tokens true cost for the adders.

Recent Price Changes

DeepSeek V4 Pro

Aug 13, 2026

$0.66 / $1.98

+102%

Qwen3.8 Max

Aug 3, 2026

$2 / $6

-20%

GPT-5.6 Terra

Jul 30, 2026

$2 / $12

-20%

GPT-5.6 Luna

Jul 30, 2026

$0.2 / $1.2

-80%

Gemini 3.6 Flash

Jul 21, 2026

$0.75 / $3.75

-50%

MiniMax M3

Jun 8, 2026

$0.6 / $2.4

+100%

DeepSeek V4 Pro

May 31, 2026

$0.435 / $0.87

-75%

Kimi K2.6

May 25, 2026

$0.73 / $3.49

-15%

GLM-5.1

May 25, 2026

$0.98 / $3.08

-35%

o3

Jun 10, 2025

$2 / $8

-80%

Understanding AI API Pricing in 2026

AI model pricing has undergone a dramatic transformation. Since GPT-4 launched in March 2023 at $30 per million input tokens, prices have fallen by over 90%: driven by competition from Anthropic, Google, and open-source challengers like DeepSeek and Meta's Llama.

Today's pricing market spans a 150x range: from Google's Gemini 2.0 Flash at $0.10/1M input tokens to Claude Opus 4 at $15/1M tokens. The key insight is that price doesn't always correlate with quality: DeepSeek V3 delivers 86% quality at just $0.27/1M tokens, while some premium models charge 50x more for marginal quality gains.

How to Optimize AI API Costs

The most effective strategy is model routing: sending simple queries to cheap, fast models and complex queries to premium models. A gateway like Swfte Connect automates this, typically reducing costs by 30-60% without sacrificing quality.

Other strategies include: using cached input pricing (offered by Google and DeepSeek), batching requests to reduce per-call overhead, and using open-source models for predictable workloads where you can self-host.

Pricing Trends to Watch

  • Price compression is accelerating: OpenAI cut GPT-5.6 Luna 80% and Terra 20% on 30 July 2026, citing inference work that reduced end-to-end serving cost by 20% and improved token-generation efficiency by more than 15%
  • DeepSeek sets the floor: V4 Pro launched at $1.74/$3.48, was discounted 75%, and had that cut made permanent on 31 May 2026 — so frontier-class quality now costs $0.435/$0.87. Every other lab is now pricing against that number
  • Open-weight pressure, not open-source pressure: the squeeze is coming from Chinese labs — DeepSeek, Moonshot, Z.ai, Alibaba, MiniMax — not Meta, which cancelled its open-frontier plans and shipped the closed Muse Spark instead
  • Reasoning premium: Models with extended thinking cost more per request because reasoning tokens bill as output. On Kimi K3 thinking cannot be disabled at any setting
  • Cached pricing is the real discount: cache reads run 10% of input at OpenAI and Google, and DeepSeek cut its cache-hit rate to a tenth of launch price on 26 April 2026 — $0.0028 per 1M tokens on V4 Flash, effectively free

Deploy a model with Swfte Connect

One gateway, every provider, per-token cost visibility. Swap models without touching your code.