✨ All tools are free — no signup required. 143 models, 22 providers, full comparison.
GPT-5.6 Sol/Terra/Luna now live, Claude Sonnet 5 pricing now permanent at $2/$10, and the best savings across 143 models from 22 providers. All tools free.
* GPT-oss is open-source (hosted inference, not OpenAI API). Cheapest API model: Gemini 2.5 Flash-Lite $0.10/1M.
The GPT-5.6 family is now live — Sol ($4/$20), Terra ($2/$12), and Luna ($0.20/$1.20), all with 1.05M context. Sol is the new flagship, matching GPT-5.5 pricing with larger context. Claude Sonnet 5 pricing is now permanent at $2/$10 — Anthropic cancelled the planned Sep 1 increase. The GPT-5.4 family remains the best value mid-tier (nano at $0.20/$1.25). GPT-5 and GPT-5 mini are deprecated (shutdown Dec 11 2026) — migrate now. Most developers are still overpaying by 60-90%.
| Model | Provider | Tier | Input / 1M | Output / 1M | Context |
|---|---|---|---|---|---|
| GPT-oss 20B | OpenAI | Budget | $0.08 | $0.35 | 128K |
| Gemini 2.5 Flash-Lite | Budget | $0.10 | $0.40 | 1M | |
| DeepSeek V4 Flash | DeepSeek | Budget | $0.22 | $0.28 | 1M |
| Mistral Small 4 | Mistral | Budget | $0.15 | $0.60 | 128K |
| Jamba Mini | AI21 | Budget | $0.20 | $0.40 | 256K |
Switch from expensive models to budget alternatives and save up to 98%.
| Switch From | Switch To | Output Savings |
|---|---|---|
|
GPT-5.5 Pro $180/1M output |
DeepSeek V4 Flash $0.28/1M output |
Save 99.8% |
|
GPT-5.5 $30/1M output |
DeepSeek V4 Flash $0.28/1M output |
Save 99.1% |
|
Claude Opus 4.8 $25/1M output |
Gemini 2.5 Flash-Lite $0.40/1M output |
Save 98.4% |
|
Claude Sonnet 5 $10/1M output intro |
Mistral Large 3 $1.50/1M output |
Save 85% |
|
GPT-5 $10/1M output |
DeepSeek V4 Pro $0.87/1M output |
Save 91.3% |
| Model | Status | Details | Replacement |
|---|---|---|---|
| GPT-5.6 Sol | ✨ New | $5.00/$30.00 per 1M tokens. 1.05M context. New OpenAI flagship, same price as GPT-5.5. | — |
| GPT-5.6 Terra | ✨ New | $2.50/$15.00 per 1M tokens. 1.05M context. Mid-tier workhorse. | — |
| GPT-5.6 Luna | ✨ New | $0.20/$1.20 per 1M tokens. 1.05M context. Budget-friendly with flagship context window. | — |
| Claude Sonnet 5 | ✨ Intro Price | $2/$10 per 1M tokens (permanent). 1M context. Best mid-tier value right now. | — |
| Mistral Small 4 | ✨ New | $0.15/$0.60 per 1M tokens. 128K context. Ultra-budget option for high-volume tasks. | — |
| Grok Build 0.1 | 💰 Repriced | xAI code API pricing updated. Now $1.00/$2.00 per 1M tokens. 256K context. | — |
| GPT-4.1 | ✨ New | $2.00/$8.00 per 1M tokens. 1M context. Replaces GPT-4o at 20% cheaper pricing. | — |
| GPT-4.1 mini | ✨ New | $0.40/$1.60 per 1M tokens. 1M context. Budget-friendly replacement for GPT-4o mini. | — |
| GPT-5 | ⚠ Deprecated | Shutdown Dec 11, 2026. $1.25/$10.00. Migrate to GPT-5.5 before deadline. | GPT-5.5 |
| GPT-5 mini | ⚠ Deprecated | Shutdown Dec 11, 2026. $0.25/$2.00. Migrate to GPT-5.4 mini before deadline. | GPT-5.4 mini |
| GPT-4.1 nano | ⚠ Deprecated | Shutdown Oct 23, 2026. $0.10/$0.40. Migrate to GPT-5.4 nano before deadline. | GPT-5.4 nano |
| Llama 4 Scout / Maverick | 🚫 Delisted | Removed from Together.ai serverless Jul 2026. May still be available via dedicated endpoints. | Llama 3.3 70B |
| Jamba 1.5 Large | ⚠ Deprecated | Superseded by Jamba 1.7 Large with same pricing. | Jamba 1.7 Large |
| Claude Sonnet 4.6 | ⚠ Deprecated | Deprecated Jun 30, 2026. Same pricing as successor. | Claude Sonnet 5 |
| Claude 4 Opus | ⚠ Deprecated | Deprecated Jun 15, 2026. Was $15/$75 — 3× more expensive than successor. | Claude Opus 4.8 |
| Claude Sonnet 4 | ⚠ Deprecated | Deprecated Jun 15, 2026. Replaced by Sonnet 4.6, then Sonnet 5. | Claude Sonnet 5 |
| DeepSeek V3.2 | ⚠ Deprecated | Removed from DeepSeek pricing page Jul 7. No longer available. | DeepSeek V4 Flash |
| DeepSeek V3 | ⚠ Deprecated | Superseded by V4 Flash at lower pricing. | DeepSeek V4 Flash |
| Llama 3.1 70B | ⚠ Deprecated | Delisted from Together.ai serverless Jul 1. $0.88/$0.88. | Llama 3.3 70B |
| Llama 3.1 8B | ⚠ Deprecated | Delisted from Together.ai serverless Jul 1. $0.10/$0.10. | Llama 3.3 70B |
| Gemini 2.0 Flash / Flash Lite | ⚠ Deprecated | Replaced by Gemini 3 Flash and 3.1 Flash-Lite. | Gemini 3 Flash |
| GPT-4o / GPT-4o mini | ⚠ Deprecated | Deprecated Apr 2025. GPT-4.1 family is 20% cheaper with 8× larger context (1M vs 128K). | GPT-4.1 / nano |
Cheapest model per provider, sorted by output cost.
| Provider | Cheapest Model | Input / 1M | Output / 1M | Context |
|---|---|---|---|---|
| OpenAI | GPT-oss 20B | $0.08 | $0.35 | 128K |
| Gemini 2.5 Flash-Lite | $0.10 | $0.40 | 1M | |
| DeepSeek | DeepSeek V4 Flash | $0.22 | $0.28 | 1M |
| Mistral | Mistral Small 4 | $0.15 | $0.60 | 128K |
| Cohere | Command R | $0.50 | $1.50 | 128K |
| xAI | Grok Build 0.1 | $1.00 | $2.00 | 256K |
| Meta | Llama 3.3 70B | $1.04 | $1.04 | 128K |
| AI21 | Jamba Mini | $0.20 | $0.40 | 256K |
| Moonshot | Kimi K2.6 | $0.95 | $4.00 | 256K |
| Anthropic | Claude Haiku 4.5 | $1.00 | $5.00 | 200K |
Use premium models (GPT-5.5, Opus 4.8) for complex reasoning and code generation. Use budget models (Flash, DeepSeek, Mistral Small) for classification, extraction, and simple Q&A. Splitting workloads across tiers typically saves 60-80% with no quality loss on routine tasks.
APIpulse's free tools cover cost calculators, model comparisons and migration guides, with every price checked against the provider's own pricing page. All tools are free.