Overall lowest
DeepSeek-V4.1 Flash
$1.19
CHF 0.956
95% below filtered median · Off-peak
Cross-provider
Compare cost, not sticker price
Shape a workload and see the lowest-cost purchasable lanes across Microsoft Foundry and direct APIs. Every result stays traceable to the complete rate table below.
Start with a pattern
Illustrative workloads
Pick a useful starting shape, then fine-tune every input below.
Cost signal
Lowest cost for this workload
Filtered median $22.89 · CHF 18.43
Foundry lowest
GPT-6 Luna
$1.25
CHF 1.00
95% below filtered median
Direct lowest
DeepSeek-V4.1 Flash
$1.19
CHF 0.956
95% below filtered median · Off-peak
Fine-tune
Workload and pricing scenario
Adjust the token shape, cache behavior, time window, and service tier. Every signal updates immediately.
Interactive
Workload cost calculator
60M
210K
90%
Share of input tokens served from cache. Applied only to models with a cache meter.
Time
Rows with a rate that changes by time of day, promo window or service tier are priced for this scenario instead of their flat listed rate; a brand-colored label under the model name names which one applies. Context size for band-priced rows uses the input-token count below.
Picking a specific hour also previews time-of-day rates that are scheduled but have not started billing yet; those rows are labelled · preview with their start instant. Now only ever shows rates that are billable at this moment.
Results
Focus the catalog
129 of 129 lanes60M in / 210K out @ 90% cache
| Shortlist | ||||||||
|---|---|---|---|---|---|---|---|---|
| DeepSeek | DeepSeek-V4.1 Flash DeepSeek direct API Off-peak | Direct API | $0.15 | $0.003 | $0.60 | $0.018 | $1.19 CHF 0.956 | |
| Qwen | Qwen3.7 Flash Model Studio (Intl) | Direct API | $0.10 | $0.01† | $0.40 | $0.019 | $1.22 CHF 0.985 | |
| OpenAI / Azure OpenAI | GPT-6 Luna OpenAI direct API | Direct API | $0.10 | $0.01 | $0.50 | $0.019 | $1.25 CHF 1.00 | |
| OpenAI / Azure OpenAI | GPT-6 Luna | Foundry ·Global | $0.10 | $0.01 | $0.50 | $0.019 | $1.25 CHF 1.00 | |
| Claude | Claude Haiku 5.5 Up To 100K Prompt | Direct API | $0.10 | $0.01 | $0.50 | $0.019 | $1.25 CHF 1.00 | |
| Claude | Claude Haiku 5.5 Up To 100K Prompt | Foundry ·Global | $0.10 | $0.01 | $0.50 | $0.019 | $1.25 CHF 1.00 | |
| OpenAI / Azure OpenAI | GPT-6 Luna | Foundry ·Data Zone | $0.11 | $0.011 | $0.55 | $0.021 | $1.37 CHF 1.10 | |
| Claude | Claude Haiku 5.5 Up To 100K Prompt | Foundry ·Data Zone | $0.11 | $0.011 | $0.55 | $0.021 | $1.37 CHF 1.10 | |
| MiniMax | MiniMax M3.1 Flash MiniMax direct API | Direct API | $0.15 | $0.015 | $0.60 | $0.029 | $1.84 CHF 1.48 | |
| Qwen | Qwen3.8 Flash Model Studio (Intl) | Direct API | $0.15 | $0.016 | $0.47 | $0.029 | $1.86 CHF 1.50 | |
| DeepSeek | DeepSeek-V4.1 Flash Direct from Azure | Foundry ·Global | $0.30 | $0.006 | $1.20 | $0.035 | $2.38 CHF 1.91 | |
| OpenAI / Azure OpenAI | GPT-6 Luna Long Context OpenAI direct API | Direct API | $0.20 | $0.02 | $0.75 | $0.038 | $2.44 CHF 1.96 |
Showing 12 of 129 lanes under the current sort
Tier badges reading Foundry · … are Microsoft Foundry deployment tiers (Global routes to any datacenter, Data Zone pins to US or EU at roughly a 10% premium, Regional pins to one region); Direct API is the model developer’s own first-party API. Blended in is the effective $/1M input paid after the cache split. * marks models with no cache meter (hit rate ignored). Daggers † /‡ mark derived / estimated rates. A brand-colored label under a model name names the variant priced under the selected scenario, when it differs from the row’s base rate. A · preview label means the opposite: that rate is scheduled but is not billing yet, and is only shown because a specific hour was picked above — the row still charges its base rate until the stated start instant. The default Now scenario never shows a preview. The ☆ control pins a lane to the shortlist tray, which ranks pinned lanes by cost for the selected workload only.