BenchLeader

LLM API pricing tracker

Input and output prices per million tokens for every model BenchLeader tracks, from first-party list prices where published and refreshed with each daily update. The ranking uses a blended price at three input tokens per output token; each model's page charts its full price history, and the last 90 days of changes are listed below the table.

As of 21 Oct 2026, the cheapest model on the pricing tracker is Gemma 4 E4B at $0.040 per million tokens (blended), followed by Llama 3.1 8B ($0.048) and Gemma 4 E2B ($0.050).

336 models · data as of 21 Oct 2026

Quality against what it costs to get there

Model size
40506070$0.0050$0.010$0.020$0.050$0.100$0.200$0.500$1.00$2.00$5.00$10.00What one full Intelligence Index run cost ($, log scale)BenchLeader IndexClaude Opus 5.5Claude Opus 5.5Gemini 4 ArgonGemini 4 ArgonGPT-6 AstraGPT-6 AstraClaude Fable 5.1Claude Fable 5.1GPT-6.1 SolGPT-6.1 SolClaude Sonnet 5.5Claude Sonnet 5.5Claude Fable 5Claude Fable 5Claude Opus 5Claude Opus 5GPT-5.6 SolGPT-5.6 SolMuse WonderMuse WonderMuse Spark 1.3GPT-6 SolGPT-5.5Kimi K3Gemini 3.8 FlashGPT-5.6 TerraMiMo-V2.6-ProGrok 4.6Grok 4.7GLM 5.3Claude Opus 4.8Qwen3.8 MaxGemini 3.5 FlashMuse Spark 1.2Qwen3.8 2.4T A95BDeepSeek V4.1 Flash 1017NEW DeepSeek V4.1 Flash 1017GLM 5.3 FlashDeepSeek V4 ProGLM 5.2Gemini 3.6 FlashDeepSeek V4.1 FlashMiniMax-M3.1-FlashNEW MiniMax-M3.1-FlashClaude Haiku 5.5Muse Spark OpenNEW Muse Spark OpenDeepSeek V4 FlashGemini 3.1 ProQwen3 7Step 5 PreviewClaude Sonnet 5Qwen3.8-Flash-NextGPT-6 LunaGPT-5.6 LunaMiMo-V2.6-FlashKimi K2.6Grok 4.5Qwen3.7 PlusDeepseek v4 Flash VisionGPT-5.5 InstantMiMo-V2.5-ProMistral Large 4GLM 5.1MiniMax M3GPT-5.4 MiniQwen3.8 27BKimi K2.7 CodeQuasar 438BNemotron 3 Ultra 550B A55BQwen3.5 397B A17BInkling SmallGemini 3.5 Flash LiteBeam 1.0NEW Beam 1.0Apodex 1.1MiMo-V2.5GPT-5.4 NanoGemini 2.5 ProQwen3.6 27BGemini 3.1 Flash LiteSolar Pro 4Qwen3.6 35B A3BNemotron 3 Super 120B A12BMistral Medium 3.5LongCat 2.0Ring-2.6-1TQwen3.5-9BGranite 4.2 30Bgpt-oss-120bNemotron 3.5 LightningTrinity LargeMercury 2DeepSeek R1Nemotron 3 Nano 30B A3BGranite 4.2 8BMistral Small 4Mistral Large 3Mercury 2.5Granite 4.2 3BGemma 4 E4Bgpt-oss-20bDeepSeek V3Solar Pro 3Llama 4 MaverickQwen3 Coder NextMistral Small 3.2Celeris-1Llama 4 ScoutMinistral 3 14BMinistral 3 8BMinistral 3 3BMistral Small 3.1
  • OpenAI13
  • Alibaba11
  • Google9
  • Mistral AI9
  • Anthropic8
  • DeepSeek7
  • Meta6
  • NVIDIA4
  • Xiaomi4
  • Zhipu AI4
  • IBM3
  • Moonshot AI3
  • SpaceXAI3
  • Inception2
  • MiniMax2
  • Upstage2
  • Ant Group1
  • Apodex1
Dashed line: the frontier. Purple-ringed points are not beaten on both axes by any other model shown. Dashed ring: new in the last seven days.

The dashed staircase is the efficiency frontier: at each price, the best index anyone reaches. A model above and left of the rest gives you more quality per dollar of real work. Size bands and run costs come from Artificial Analysis, which is the only source that publishes what a fixed batch of work actually cost.

336 of 336
#
1o1-pro$262.50$150.00$600.00
2GPT-5.5 Proxhigh$67.50$30.00$180.00
3GPT-5.4 Pro$67.50$30.00$180.00
4GPT-5.2 Proxhigh$57.75$21.00$168.00
5GPT-5 Pro$41.25$15.00$120.00
6GPT-4$37.50$30.00$60.00
7GPT 4 0613$37.50$30.00$60.00
8o3-prohigh$35.00$20.00$80.00
9Claude Opus 4.1thinking$30.00$15.00$75.00
10Claude Opus 4thinking$30.00$15.00$75.00
11Claude 3 Opus$30.00$15.00$75.00
12o1$26.25$15.00$60.00
13GPT-6 Astramax$20.00$10.00$50.00
14Claude Fable 5.1max$20.00$10.00$50.00
15Claude Fable 5max$20.00$10.00$50.00
16GPT-6 Astra Promax$20.00$10.00$50.00
17GPT-4 Turbo$15.00$10.00$30.00
18GPT-5.5xhigh$11.25$5.00$30.00
19GPT-5.5 Instant$11.25$5.00$30.00
20Claude Opus 5max$10.00$5.00$25.00
21Claude Opus 4.7$10.00$5.00$25.00
22Claude Opus 4.8max$10.00$5.00$25.00
23Claude Opus 4.6high$10.00$5.00$25.00
24Claude Opus 4.5thinking$10.00$5.00$25.00
25Step 2$8.02$5.21$16.44
26Claude Opus 5.5max$8.00$4.00$20.00
27GPT-5.6 Solmax$8.00$4.00$20.00
28Muse Wondermax$8.00$4.00$20.00
29GPT-5.6 Sol Promax$8.00$4.00$20.00
30GPT-5.6$8.00$4.00$20.00
31Kimi K3max$6.00$3.00$15.00
32Claude Sonnet 4.5high$6.00$3.00$15.00
33Grok 4$6.00$3.00$15.00
34Claude Sonnet 4.6$6.00$3.00$15.00
35Claude Sonnet 4thinking$6.00$3.00$15.00
36Grok 3$6.00$3.00$15.00
37Claude 3.5 Sonnet$6.00$3.00$15.00
38Mistral Large 24.02$6.00$4.00$12.00
39Ember-1$6.00$3.00$15.00
40Sonar Pro$6.00$3.00$15.00
41GPT-5.4xhigh$5.63$2.50$15.00
42Nova Premier$5.00$2.50$12.50
43GPT-5.3-Codexxhigh$4.81$1.75$14.00
44GPT-5.2xhigh$4.81$1.75$14.00
45GPT-5.2-Codexxhigh$4.81$1.75$14.00
46GPT-5.3 Chat$4.81$1.75$14.00
47GPT-5.6 Terramax$4.50$2.00$12.00
48Gemini 3.1 Pro$4.50$2.00$12.00
49Gemini Pro Latest$4.50$2.00$12.00
50Gemini 3.1 Pro Customtools$4.50$2.00$12.00
51GPT-4o$4.38$2.50$10.00
52Command A$4.38$2.50$10.00
53Command R Plus 08 2024$4.38$2.50$10.00
54Command A Plus 05 2026$4.38$2.50$10.00
55Command A Reasoningthinking$4.38$2.50$10.00
56Gemini 4 Argonhigh$4.00$2.00$10.00
57GPT-6.1 Solmax$4.00$2.00$10.00
58Claude Sonnet 5.5max$4.00$2.00$10.00
59GPT-6 Solmax$4.00$2.00$10.00
60Claude Sonnet 5high$4.00$2.00$10.00

Price changes in the last 90 days

Every list-price change over 1% (blended) up to 21 Oct 2026, newest first. Each model’s page charts its full price history.

ModelDateBefore (in / out)After (in / out)Blended change
DeepSeek V4.1 Flash 101719 Oct 2026$0.090 / $0.550$0.049 / $0.600-9%
DeepSeek V4 Flash17 Oct 2026$0.024 / $0.405$0.018 / $0.036-81%
DeepSeek V4 Pro17 Oct 2026$0.255 / $0.510$0.245 / $0.490-4%
DeepSeek V4 Pro16 Oct 2026$1.69 / $3.38$0.255 / $0.510-85%
DeepSeek V4 Flash16 Oct 2026$0.045 / $1.15$0.024 / $0.405-63%
Kimi K2.616 Oct 2026$0.950 / $4.00$0.432 / $1.82-54%
GLM 5.3 Flash16 Oct 2026$0.036 / $0.900$0.028 / $0.446-48%
GLM 5.116 Oct 2026$1.40 / $4.40$0.965 / $3.03-31%
Gemma 4 26B A4B16 Oct 2026$0.076 / $0.255$0.068 / $0.225-12%
GLM 5.216 Oct 2026$0.613 / $1.93$0.556 / $1.75-9%
DeepSeek V4.1 Flash16 Oct 2026$0.011 / $0.800$0.200 / $0.600+44%
DeepSeek V4.1 Flash15 Oct 2026$0.016 / $1.20$0.011 / $0.800-33%
Kimi K315 Oct 2026$0.690 / $17.50$1.00 / $14.90-9%
DeepSeek V4 Flash15 Oct 2026$0.037 / $1.15$0.045 / $1.15+2%
Qwen3.8 27B15 Oct 2026$0.040 / $2.30$0.200 / $6.00+173%
GLM 5.214 Oct 2026$0.643 / $2.02$0.613 / $1.93-5%
Kimi K314 Oct 2026$1.00 / $14.90$0.690 / $17.50+9%
GLM 5.314 Oct 2026$0.055 / $5.00$0.087 / $6.00+21%
Gemma 3 27B14 Oct 2026$0.070 / $0.275$0.080 / $0.350+22%
GPT-5.6 Sol14 Oct 2026$2.00 / $10.00$4.00 / $20.00+100%
DeepSeek V4 Flash13 Oct 2026$0.018 / $1.54$0.037 / $1.15-21%
GLM 5.3 Flash13 Oct 2026$0.023 / $1.20$0.036 / $0.900-21%
Gemma 3 27B13 Oct 2026$0.080 / $0.300$0.070 / $0.275-10%
GLM 5.113 Oct 2026$0.965 / $3.03$1.40 / $4.40+45%
Kimi K2.613 Oct 2026$0.432 / $1.82$0.950 / $4.00+120%
DeepSeek V4 Pro13 Oct 2026$0.414 / $0.828$1.69 / $3.38+308%
DeepSeek V4 Pro12 Oct 2026$1.69 / $3.38$0.414 / $0.828-76%
Kimi K2.612 Oct 2026$0.950 / $4.00$0.432 / $1.82-54%
GLM 5.112 Oct 2026$1.40 / $4.40$0.965 / $3.03-31%
GLM 5.312 Oct 2026$0.250 / $5.00$0.055 / $5.00-10%
Gemma 4 26B A4B12 Oct 2026$0.068 / $0.225$0.076 / $0.255+13%
Kimi K312 Oct 2026$0.288 / $14.90$1.00 / $14.90+14%
GLM 5.212 Oct 2026$0.556 / $1.75$0.643 / $2.02+16%
GLM 5.3 Flash12 Oct 2026$0.032 / $0.720$0.023 / $1.20+56%
MiMo-V2.6-Flash11 Oct 2026$0.260 / $0.410$0.190 / $0.290-28%
GLM 5.311 Oct 2026$0.393 / $6.00$0.250 / $5.00-20%
DeepSeek V4 Flash11 Oct 2026$0.033 / $1.15$0.018 / $1.54+27%
GLM 5.3 Flash11 Oct 2026$0.044 / $0.450$0.032 / $0.720+40%
Kimi K2.611 Oct 2026$0.434 / $1.83$0.950 / $4.00+119%
DeepSeek V4 Pro11 Oct 2026$0.171 / $0.341$1.69 / $3.38+890%

And 306 earlier changes, on each model’s page.

Cite as: BenchLeader, “LLM API pricing tracker”, https://www.benchleader.com/leaderboards/cost, data as of 21 Oct 2026.