OOpenAI
pricingOpenAI cuts GPT-5 input by 17%, expands cached input discount
Standard input drops from $3.00 to $2.50 per 1M. Cached input now $0.625 per 1M, a 50% reduction.
2026-05-15OpenAI
Market preview · —
Track AI model pricing like a market.
Tokentrendr monitors token pricing across top AI platforms and shows where costs are rising, falling, and creating savings opportunities.
Green down = price drop, savings Red up = price increase
Market snapshotUSD / 1M tokens
OA
GPT-5
OpenAI
$2.50
-16.7%
AN
Claude 4.5 Sonnet
Anthropic
$3.00
-8.3%
GG
Gemini 2.5 Pro
Google
$1.25
-10.7%
xA
Grok 4
xAI
$3.00
+7.1%
MS
Mistral Large 3
Mistral
$2.00
-20.0%
DPSDeepSeek R1$0.550-45.0%OAIGPT-5 mini$0.250-28.6%GGLGemini 2.5 Flash$0.150-25.0%COHCommand R$0.150-25.0%DPSDeepSeek V3.2$0.270-22.9%MSTMistral Large 3$2.00-20.0%OAIGPT-5$2.50-16.7%COHCommand R+$2.50-16.7%ANTClaude 4.5 Haiku$0.800+14.3%xAIGrok 4 mini$0.300-14.3%GGLGemini 2.5 Pro$1.25-10.7%MTALlama 4 405B$2.70-10.0%DPSDeepSeek R1$0.550-45.0%OAIGPT-5 mini$0.250-28.6%GGLGemini 2.5 Flash$0.150-25.0%COHCommand R$0.150-25.0%DPSDeepSeek V3.2$0.270-22.9%MSTMistral Large 3$2.00-20.0%OAIGPT-5$2.50-16.7%COHCommand R+$2.50-16.7%ANTClaude 4.5 Haiku$0.800+14.3%xAIGrok 4 mini$0.300-14.3%GGLGemini 2.5 Pro$1.25-10.7%MTALlama 4 405B$2.70-10.0%
Market overview
Today on the board
Providers tracked
9
Models tracked
20
Biggest drop
-45.0%
DeepSeek R1
Biggest increase
+14.3%
Claude 4.5 Haiku
Average move
-11.1%
12 down · 2 up
Last updated
—
Today's flow
Biggest movers
The largest pricing changes in the selected range.
Snapshot
Top providers
Pricing comparison
All models, all providers
Filterable comparison across input, output, cached input, context window, and recent movement.
| Provider | Cached | Trend | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Gemini 2.5 Flash-Lite budget | GGoogle | $0.075 | $0.300 | — | 1000K | 0.0% | View | ||
| Gemini 2.5 Flash fast | GGoogle | $0.150 | $0.600 | $0.037 | 1000K | -25.0% | View | ||
| Command R fast | CCohere | $0.150 | $0.600 | — | 128K | -25.0% | View | ||
| GPT-5 mini fast | OOpenAI | $0.250 | $2.00 | $0.050 | 400K | -28.6% | View | ||
| DeepSeek V3.2 frontier | DDeepSeek | $0.270 | $1.10 | $0.070 | 128K | -22.9% | View | ||
| Grok 4 mini fast | xxAI | $0.300 | $1.50 | — | 131K | -14.3% | View | ||
| Codestral 2 coding | MMistral | $0.400 | $1.20 | — | 256K | 0.0% | View | ||
| DeepSeek R1 reasoning | DDeepSeek | $0.550 | $2.19 | — | 64K | -45.0% | View | ||
| Llama 4 70B fast | MMeta | $0.590 | $0.790 | — | 128K | 0.0% | View | ||
| Claude 4.5 Haiku fast | AAnthropic | $0.800 | $4.00 | $0.080 | 200K | +14.3% | View | ||
| Sonar Large multimodal | PPerplexity | $1.00 | $1.00 | — | 127K | 0.0% | View | ||
| Gemini 2.5 Pro frontier | GGoogle | $1.25 | $10.00 | $0.310 | 2000K | -10.7% | View | ||
| Mistral Large 3 frontier | MMistral | $2.00 | $6.00 | — | 128K | -20.0% | View | ||
| GPT-5 frontier | OOpenAI | $2.50 | $10.00 | $0.625 | 400K | -16.7% | View | ||
| Command R+ frontier | CCohere | $2.50 | $10.00 | — | 128K | -16.7% | View | ||
| Llama 4 405B frontier | MMeta | $2.70 | $2.70 | — | 128K | -10.0% | View | ||
| Claude 4.5 Sonnet frontier | AAnthropic | $3.00 | $15.00 | $0.300 | 200K | -8.3% | View | ||
| Grok 4 frontier | xxAI | $3.00 | $15.00 | — | 256K | +7.1% | View | ||
| o4 reasoning | OOpenAI | $5.00 | $20.00 | — | 200K | 0.0% | View | ||
| Claude 4.5 Opus reasoning | AAnthropic | $15.00 | $75.00 | $1.50 | 200K | 0.0% | View |
Recent updates
Pricing news
DDeepSeek
pricingDeepSeek V3.2 launches with sparse attention, 23% cheaper than V3.1
Long-context inputs become materially cheaper via sparse attention. R1 also receives a permanent price cut.
2026-05-15DeepSeek
AAnthropic
apiAnthropic expands prompt caching to all Claude 4.5 tiers
Cached input now 1/10th the price of standard input across Sonnet, Haiku, and Opus.
2026-05-14Anthropic
GGoogle
pricingGemini 2.5 Flash price drops 25% across Vertex and AI Studio
Standard input now $0.15 per 1M tokens. Cached input remains 25% of standard.
2026-05-12Google
AAnthropic
pricingClaude 4.5 Haiku input rises 14% alongside context window upgrade
Haiku now supports a 200K context window. The price increase reflects the expanded capability.
2026-04-01Anthropic
xxAI
pricingGrok 4 standard input increases 7%
xAI cites realtime context infrastructure costs. Grok 4 mini remains unchanged.
2026-04-12xAI
Savings library
Tips to cut token spend
CachingBeginnerHigh impact
Use prompt caching for repeated system prompts
Long, repeated system or tool-definition blocks can cost 90% less when cached. Restructure prompts so static content comes first.
OpenAIAnthropicGoogleDeepSeek
BatchingBeginnerHigh impact
Route async workloads to Batch APIs
Most providers offer 50% off when results can return within 24 hours. Ideal for evals, backfills, and enrichment.
OpenAIAnthropicGoogle
RoutingIntermediateHigh impact
Route by task complexity, not by default to your largest model
Classify intent with a small model, then escalate. A simple router can cut spend 60%+ on mixed workloads.
Prompt designBeginnerMedium impact
Cap output tokens deliberately
Output tokens are typically 4–5x the price of input. Set max_tokens, request concise schemas, and prefer structured outputs.
ContextIntermediateMedium impact
Prune context aggressively
Drop turn-history older than what the model needs. Summarize older state into a short memory block.
Prompt designBeginnerMedium impact
Use structured outputs over freeform JSON parsing
Native structured outputs reduce retries and wasted tokens from malformed JSON.
OpenAIAnthropicGoogle
Methodology
How Tokentrendr measures the market
Prices are normalized to USD per 1M tokens. We track input, output, cached input, and batch discounts from official pricing pages and provider changelogs. Movement is calculated against the previous published price.
Read the full methodology- Normalized unitsPer 1M tokens, USD
- SourcesOfficial pricing pages and changelogs
- Update cadenceEvery 6 hours via Firecrawl scrape
- Direction colorsGreen = down (good), Red = up (bad)