Best LLM APIs with Prompt Caching
LLM APIs that support prompt caching, ranked by cached-input price. Cache hits cost 50–99% less than fresh input — the biggest lever for cutting cost on repeated system prompts, RAG context, and long conversations.
MODELPRICEWATCH.COM · 2026-11-14
Cost calculator for this use casemonthly cost, top 3 models
- 1 GLM-4.6V-FlashX
- $—
- 2 Muse Ember
- $—
- 3 DeepSeek V4.1 Flash 1017
- $—
Full ranking — top 20 models
list prices, USD per 1M tokens| Model | Cached input | Blended* | Input | Output | Context | Provider |
|---|---|---|---|---|---|---|
| 1GLM-4.6V-FlashX | $0.004 | $0.130 | $0.040 | $0.400 | 128K | Z.AI |
| 2Muse Ember | $0.004 | $0.200 | $0.100 | $0.500 | 1M | Meta |
| 3DeepSeek V4.1 Flash 1017 | $0.006 | $0.525 | $0.300 | $1.20 | 1M | DeepSeek |
| 4LongCat-2.5 | $0.006 | $0.525 | $0.300 | $1.20 | 1M | Meituan |
| 5GPT-6 Luna | $0.010 | $0.200 | $0.100 | $0.500 | 1M | OpenAI |
| 6GLM-4.7-FlashX | $0.010 | $0.153 | $0.070 | $0.400 | 128K | Z.AI |
| 7Solar Mini 4 | $0.010 | $0.175 | $0.100 | $0.400 | 524K | Upstage |
| 8GPT-OSS 120B | $0.015 | $0.262 | $0.150 | $0.600 | 128K | Fireworks |
| 9MiniMax-M3.1-Flash | $0.015 | $0.262 | $0.150 | $0.600 | 1M | MiniMax |
| 10Qwen3.8-Omni-Flash | $0.016 | $0.230 | $0.150 | $0.470 | 1M | Alibaba |
| 11Step 3.5 Flash | $0.020 | $0.150 | $0.100 | $0.300 | 262K | StepFun |
| 12LFM3-1.7B | $0.020 | $0.110 | $0.080 | $0.200 | 262K | Liquid AI |
| 13LFM2.5-24B-A2B | $0.025 | $0.212 | $0.050 | $0.700 | 131K | Liquid AI |
| 14GLM-4.5-Air | $0.030 | $0.425 | $0.200 | $1.10 | 128K | Z.AI |
| 15GLM-5.3-Flash | $0.030 | $0.237 | $0.150 | $0.500 | 1M | Z.AI |
| 16NVIDIA Nemotron 3.5 Lightning | $0.030 | $0.085 | $0.060 | $0.160 | 256K | DeepInfra |
| 17Inkling 2 Small | $0.030 | $0.475 | $0.300 | $1.00 | 262K | Thinking Machines |
| 18KAT-Coder-Air V2.5 | $0.030 | $0.259 | $0.148 | $0.593 | 256K | Kwaipilot |
| 19Pareto | $0.030 | $1.40 | $0.800 | $3.20 | 1M | Unbiased |
| 20Particle-1 | $0.030 | $0.438 | $0.250 | $1.00 | 1M | Particle AI |
Recent price movement in this ranking
price-only deltas · logged by the daily scan4 of the top 20 prompt caching models have re-priced since we began tracking · last checked .
| Model | Changes | Latest move | Provider |
|---|---|---|---|
| Solar Mini 4 | 1 | price rise on : $0.050/$0.200 → $0.100/$0.400 /Mtok | Upstage |
| NVIDIA Nemotron 3.5 Lightning | 1 | price cut on : $0.080/$0.200 → $0.060/$0.160 /Mtok | DeepInfra |
| Pareto | 1 | price cut on : $2.50/$7.50 → $0.800/$3.20 /Mtok | Unbiased |
| GLM-5.3-Flash | 1 | price rise on : $0.075/$0.250 → $0.150/$0.500 /Mtok | Z.AI |
Most recently, Solar Mini 4 raised its price on — a sign pricing in this category is still moving, so re-check before committing to a long-term choice. See all recent price moves →
How models are selected
Models offering prompt caching, sorted by cached-input price per million tokens (cheapest cache reads first).
Prices are per million tokens (Mtok), sourced directly from — and linked to — official provider pricing pages. "Blended cost" is (3×input + 1×output) ÷ 4 — weighted toward input because real workloads read far more tokens than they generate.