AI pricing inside the Mindcrew tool network
Compare AI model costs here, or jump to calculators, code formatting, and Markdown conversion across the Mindcrew browser tools.
Pricing
Compare tracked AI models by benchmark workload, current pricing units, context windows, quality tiers, and source records.
Popular comparisons
Original analysis paired with a live pricing snapshot for common AI buying decisions — not just a reproduced price table.
Head-to-head comparisons
How OpenAI and Anthropic API pricing compares at a moderate production workload, tier for tier.
Open GPT vs Claude pricingA flat monthly seat price is only half the comparison — here is the other half.
Open ChatGPT Plus vs Claude ProSeat pricing for a 12-developer team, plus what the price alone won't tell you about either tool.
Open Cursor vs GitHub CopilotThe GPT vs Claude comparison changes shape once volume and discount eligibility enter the picture.
Open OpenAI vs Anthropic API pricing at scaleGoogle and Anthropic API pricing, tier for tier, at a moderate production workload.
Open Gemini vs Claude pricingFor teams shopping primarily on price, here is what the gap actually looks like at balanced and budget tiers.
Open Mistral vs OpenAI pricingDeepSeek built its reputation on being dramatically cheaper than the US labs — here is the actual multiple, tier for tier.
Open DeepSeek vs OpenAI pricingDeepSeek's reasoning tier against Claude, plus the balanced-tier gap most buyers actually ask about.
Open DeepSeek vs Claude pricingxAI's frontier and balanced tiers against OpenAI's, at a moderate production workload.
Open Grok vs GPT pricingTwo agentic-editor rivals, priced at a 12-developer team size — plus Windsurf's premium tier most comparisons skip.
Open Windsurf vs CursorJetBrains' entry tier undercuts Copilot by half — here is what that gap and the top tier actually look like at team size.
Open JetBrains AI vs CopilotBest & cheapest picks
Budget and fast-tier coding-capable models, ranked by live monthly cost — and what you give up to get there.
Open Cheapest LLM API for codingBalanced-tier models at a realistic heavy-chat workload, ranked on cost with quality tier shown alongside.
Open Best value LLM for chatBudget-tier models ranked at 250M tokens a month, with committed-use and volume discounts applied where offered.
Open Cheapest LLM API for high volumeSix coding-assistant subscriptions priced at a 12-developer team size, ranked by monthly seat cost.
Open Best AI coding assistant for teamsThe same open-weight models, hosted by different providers — sometimes tied, sometimes 4x apart.
Open Cheapest open-model inference APIOpenAI, Anthropic, Google, and xAI's top-tier models, ranked by live monthly cost at the same workload.
Open Frontier LLM pricing comparedIndustry benchmark map
Solid points use the model's native unit for the selected benchmark. Muted points use a translated spend lens so subscription, token, and media pricing can still be compared.
Pricing data collected latest verified 7/16/2026, 12:00:00 AM; effective 7/14/2026 - 7/16/2026.
Frequent personal and work chat across writing, information seeking, and decisions. Moderate production API workload: 0.8M input and 0.4M output tokens per month.
All⌄
Quality view plots estimated monthly cost against the model quality score. The score uses catalogue qualityScore when present, otherwise tier fallbacks: frontier 90, balanced 70, unrated 55, budget 45.
Hold Ctrl and drag a rectangle over the chart to zoom into that area.
AI21 Labs Jamba Large 1.7: $4.80 per month, context 256,000. Refreshing.
AI21 Labs Jamba 1.5 Mini: $0.32 per month, context unknown. Refreshing.
Alibaba Cloud (Qwen) Qwen-Max: $5.00 per month, context unknown. Refreshing.
Alibaba Cloud (Qwen) Qwen-Plus: $0.96 per month, context unknown. Refreshing.
Alibaba Cloud (Qwen) Qwen-Turbo: $0.12 per month, context unknown. Refreshing.
Amazon Bedrock Amazon Nova Pro: $1.92 per month, context unknown. Refreshing.
Amazon Bedrock Amazon Nova Lite: $0.144 per month, context unknown. Refreshing.
Amazon Bedrock Amazon Nova Micro: $0.084 per month, context unknown. Refreshing.
Amazon Q Developer Q Developer Pro: $19.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Anthropic Claude Opus 4.8: $14.00 per month, context 1,000,000. Refreshing.
Anthropic Claude Sonnet 5: $5.60 per month, context 1,000,000. Refreshing.
Anthropic Claude Sonnet 4.6: $8.40 per month, context 1,000,000. Refreshing.
Anthropic Claude Haiku 4.5: $2.80 per month, context unknown. Refreshing.
ChatGPT (OpenAI) ChatGPT Plus: $20.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
ChatGPT (OpenAI) ChatGPT Pro: $200.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Claude (Anthropic) Claude Pro: $20.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Claude (Anthropic) Claude Max: $100.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Cohere Command R+: $6.00 per month, context 128,000. Refreshing.
Cursor (Anysphere) Cursor Pro: $20.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
DeepSeek DeepSeek V4 Flash: $0.224 per month, context 1,000,000. Refreshing.
DeepSeek DeepSeek V4 Pro: $0.696 per month, context 1,000,000. Refreshing.
DeepSeek DeepSeek Chat V3.2: $0.392 per month, context unknown. Refreshing.
DeepSeek DeepSeek R1: $1.316 per month, context unknown. Refreshing.
Fireworks AI Kimi K2.6: $2.36 per month, context unknown. Refreshing.
Fireworks AI DeepSeek V4 Pro: $2.784 per month, context unknown. Refreshing.
Fireworks AI GPT-OSS 120B: $0.36 per month, context unknown. Refreshing.
GitHub Copilot Copilot Pro: $20.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Google Gemini 3 Pro: $6.40 per month, context 1,000,000. Refreshing.
Google Gemini 3 Flash: $1.60 per month, context unknown. Refreshing.
Google Antigravity Antigravity Pro: $20.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Google Antigravity Antigravity Ultra: $100.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Google Antigravity Antigravity Ultra Max: $200.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Groq Llama 3.3 70B Versatile: $0.788 per month, context unknown. Refreshing.
Groq GPT-OSS 120B: $0.36 per month, context unknown. Refreshing.
Groq Llama 3.1 8B Instant: $0.072 per month, context unknown. Refreshing.
JetBrains AI Assistant AI Pro: $10.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
JetBrains AI Assistant AI Ultimate: $30.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Kiro (AWS) Kiro Pro: $20.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Kiro (AWS) Kiro Pro+: $40.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Kiro (AWS) Kiro Pro Max: $100.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Kiro (AWS) Kiro Power: $200.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Mistral Mistral Large 2: $4.00 per month, context unknown. Refreshing.
Mistral Mistral Small: $0.40 per month, context unknown. Refreshing.
Mistral Ministral 3B: $0.048 per month, context unknown. Refreshing.
Moonshot AI (Kimi) Kimi K2.6: $2.36 per month, context unknown. Refreshing.
Moonshot AI (Kimi) Kimi K2.5: $1.68 per month, context unknown. Refreshing.
OpenAI GPT-5.6 Sol: $16.00 per month, context 1,050,000. Refreshing.
OpenAI GPT-5.6 Terra: $8.00 per month, context 1,050,000. Refreshing.
OpenAI GPT-5.6 Luna: $3.20 per month, context 1,050,000. Refreshing.
OpenAI GPT-5.5: $16.00 per month, context 400,000. Refreshing.
OpenAI GPT-5.3 Codex: $7.00 per month, context unknown. Refreshing.
OpenAI Codex Codex Go: $8.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
OpenAI Codex Codex Plus: $20.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
OpenAI Codex Codex Pro: $100.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Perplexity Sonar: $1.20 per month, context unknown. Refreshing.
Perplexity Sonar Pro: $8.40 per month, context unknown. Refreshing.
Perplexity Sonar Reasoning Pro: $4.80 per month, context unknown. Refreshing.
Replit Agent Replit Core: $25.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Replit Agent Replit Pro: $100.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Sourcegraph Cody Cody Enterprise: $59.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Tabnine Code Assistant: $39.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Tabnine Agentic Platform: $59.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Together AI DeepSeek V4 Pro: $2.784 per month, context unknown. Refreshing.
Together AI Qwen3.7 Plus: $0.768 per month, context unknown. Refreshing.
Together AI Kimi K2.7 Code: $2.36 per month, context unknown. Refreshing.
Windsurf (Cognition) Windsurf Pro: $20.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
Windsurf (Cognition) Windsurf Max: $200.00 per month, context unknown. Translated lens: Access-floor lens: 1 paid seat plotted as the subscription footprint for operating this workload, not as equivalent token throughput. Refreshing.
xAI Grok 4.5: $4.00 per month, context 500,000. Refreshing.
xAI Grok 4.3: $2.00 per month, context unknown. Refreshing.
xAI Grok 4.1 Fast: $0.36 per month, context 2,000,000. Refreshing.
xAI Grok 4: $8.40 per month, context 256,000. Refreshing.
xAI Grok Code Fast 1: $0.76 per month, context 256,000. Refreshing.
11 models still lack enough pricing fields for this benchmark lens.
Chart labels and measures
20 × $0.06 + 4 × $0.24 = $2.16/month.| Price options | Channel | Status | Notes | Discounts | Source | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| AI21 Labs | Jamba Large 1.7 jamba-large-1.7 Compute ROI → | Llm Api | Text | Balanced | 256,000 tokens | - | - | Direct | Priced | Hybrid Mamba-Transformer architecture; released 2025-08-08. | Per 1m Tokens | $2.00 | $8.00 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| AI21 Labs | Jamba 1.5 Mini jamba-1.5-mini Compute ROI → | Llm Api | Text | Budget | Unknown | - | - | Direct | Priced | Smaller/cheaper Jamba tier. | Per 1m Tokens | $0.20 | $0.40 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Alibaba Cloud (Qwen) | Qwen-Max qwen-max Compute ROI → | Llm Api | Text | Frontier | Unknown | - | - | Direct | Priced | International (Singapore) pricing; China/Beijing region is substantially cheaper. | Per 1m Tokens | $2.50 | $7.50 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Alibaba Cloud (Qwen) | Qwen-Plus qwen-plus Compute ROI → | Llm Api | Text | Balanced | Unknown | - | - | Direct | Priced | International pricing, 0-256K context tier. | Per 1m Tokens | $0.40 | $1.60 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Alibaba Cloud (Qwen) | Qwen-Turbo qwen-turbo Compute ROI → | Llm Api | Text | Budget | Unknown | - | - | Direct | Priced | Cheapest Qwen tier, international pricing. | Per 1m Tokens | $0.05 | $0.20 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Amazon Bedrock | Amazon Nova Pro amazon-nova-pro Compute ROI → | Cloud Hosting | Multimodal | Balanced | Unknown | - | Batch: 50% | Aws Bedrock | Priced | Via Bedrock. Batch -50%, cache up to -90%. | Per 1m Tokens | $0.80 | $3.20 | - | USD | Batch 50% | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Amazon Bedrock | Amazon Nova Lite amazon-nova-lite Compute ROI → | Cloud Hosting | Multimodal | Budget | Unknown | - | Batch: 50% | Aws Bedrock | Priced | Via Bedrock. | Per 1m Tokens | $0.06 | $0.24 | - | USD | Batch 50% | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Amazon Bedrock | Amazon Nova Micro amazon-nova-micro Compute ROI → | Cloud Hosting | Text | Budget | Unknown | - | Batch: 50% | Aws Bedrock | Priced | Cheapest Nova; via Bedrock. | Per 1m Tokens | $0.035 | $0.14 | - | USD | Batch 50% | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Amazon Bedrock | Amazon Nova Premier amazon-nova-premier Compute ROI → | Cloud Hosting | Multimodal | Frontier | Unknown | - | Batch: 50% | Aws Bedrock | Needs Verification | Top Nova tier; pricing to confirm. | No current price | - | - | - | USD | Batch 50% | Missing | Missing | Missing |
| Amazon Q Developer | Q Developer Pro q-developer-pro Compute ROI → | Coding Tool | Code | Balanced | Unknown | - | - | Direct | Priced | Single paid tier; 1,000 agentic requests/mo, 4,000 LOC/mo. Free tier also exists (unlimited completions, 50 agent chats/mo). | Per Seat Month | - | - | $19.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Anthropic | Claude Opus 4.8 claude-opus-4-8 Compute ROI → | Llm Api | Multimodal | Frontier | 1,000,000 tokens | - | Cache input: $0.50 / 1M Batch: 50% | Direct | Priced | Cache -90%, batch -50%. cachedInput approx (-90% of input). | Per 1m Tokens | $5.00 | $25.00 | - | USD | Cache 90%, Batch 50% | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Anthropic | Claude Sonnet 5 claude-sonnet-5 Compute ROI → | Llm Api | Multimodal | Balanced | 1,000,000 tokens | - | Batch: 50% | Direct | Priced | INTRO pricing through 2026-08-31; reverts to $3/$15 on 2026-09-01. | Per 1m Tokens | $2.00 | $10.00 | - | USD | Batch 50% | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Anthropic | Claude Sonnet 4.6 claude-sonnet-4-6 Compute ROI → | Llm Api | Multimodal | Balanced | 1,000,000 tokens | - | Batch: 50% | Direct | Priced | Standard Sonnet pricing. | Per 1m Tokens | $3.00 | $15.00 | - | USD | Batch 50% | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Anthropic | Claude Haiku 4.5 claude-haiku-4-5 Compute ROI → | Llm Api | Text | Fast | Unknown | - | Batch: 50% | Direct | Priced | Fast/cheap tier. | Per 1m Tokens | $1.00 | $5.00 | - | USD | Batch 50% | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Anthropic | Claude Fable 5 claude-fable-5 Compute ROI → | Llm Api | Multimodal | Balanced | 1,000,000 tokens | - | Batch: 50% | Direct | Needs Verification | Full 1M context at standard pricing; per-token price to confirm. | No current price | - | - | - | USD | Batch 50% | Missing | Missing | Missing |
| Anthropic | Claude Mythos 5 claude-mythos-5 Compute ROI → | Llm Api | Multimodal | Frontier | 1,000,000 tokens | - | Batch: 50% | Direct | Needs Verification | Newer line; pricing to confirm. | No current price | - | - | - | USD | Batch 50% | Missing | Missing | Missing |
| ChatGPT (OpenAI) | ChatGPT Plus chatgpt-plus Compute ROI → | Coding Tool | Text | Balanced | Unknown | - | - | Direct | Priced | - | Per Seat Month | - | - | $20.00 | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| ChatGPT (OpenAI) | ChatGPT Pro chatgpt-pro Compute ROI → | Coding Tool | Text | Frontier | Unknown | - | - | Direct | Priced | - | Per Seat Month | - | - | $200.00 | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Claude (Anthropic) | Claude Pro claude-pro Compute ROI → | Coding Tool | Code | Balanced | Unknown | - | - | Direct | Priced | - | Per Seat Month | - | - | $20.00 | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Claude (Anthropic) | Claude Max claude-max Compute ROI → | Coding Tool | Code | Frontier | Unknown | - | - | Direct | Priced | - | Per Seat Month | - | - | $100.00 | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Cohere | Command R+ command-r-plus Compute ROI → | Llm Api | Text | Balanced | 128,000 tokens | - | - | Direct | Priced | Flagship chat model. | Per 1m Tokens | $2.50 | $10.00 | - | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Cohere | Command A command-a Compute ROI → | Llm Api | Text | Balanced | Unknown | - | - | Direct | Needs Verification | Newer/cheaper chat tier; pricing to confirm. | No current price | - | - | - | USD | - | Missing | Missing | Missing |
| Cursor (Anysphere) | Cursor Pro cursor-pro Compute ROI → | Coding Tool | Code | Balanced | Unknown | - | - | Direct | Priced | - | Per Seat Month | - | - | $20.00 | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| DeepSeek | DeepSeek V4 Flash v4-flash Compute ROI → | Llm Api | Text | Budget | 1,000,000 tokens | - | Cache input: $0.0028 / 1M | Direct | Priced | Cache hits ~-98%. | Per 1m Tokens | $0.14 | $0.28 | - | USD | Cache 98% | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| DeepSeek | DeepSeek V4 Pro v4-pro Compute ROI → | Llm Api | Text | Balanced | 1,000,000 tokens | - | - | Direct | Priced | Flagship. | Per 1m Tokens | $0.435 | $0.87 | - | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| DeepSeek | DeepSeek Chat V3.2 chat-v3.2 Compute ROI → | Llm Api | Text | Budget | Unknown | - | - | Direct | Priced | Prior chat model. | Per 1m Tokens | $0.28 | $0.42 | - | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| DeepSeek | DeepSeek R1 r1 Compute ROI → | Llm Api | Text | Reasoning | Unknown | - | - | Direct | Priced | Reasoning model. | Per 1m Tokens | $0.55 | $2.19 | - | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Fireworks AI | Kimi K2.6 kimi-k2.6 Compute ROI → | Llm Api | Text | Balanced | Unknown | - | - | Direct | Priced | Hosted open-weight model; standard serverless tier. | Per 1m Tokens | $0.95 | $4.00 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Fireworks AI | DeepSeek V4 Pro deepseek-v4-pro Compute ROI → | Llm Api | Text | Balanced | Unknown | - | - | Direct | Priced | Hosted open-weight model; standard serverless tier. | Per 1m Tokens | $1.74 | $3.48 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Fireworks AI | GPT-OSS 120B gpt-oss-120b Compute ROI → | Llm Api | Text | Budget | Unknown | - | - | Direct | Priced | OpenAI open-weight model hosted serverless. | Per 1m Tokens | $0.15 | $0.60 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| GitHub Copilot | Copilot Pro copilot-pro Compute ROI → | Coding Tool | Code | Balanced | Unknown | - | - | Direct | Priced | - | Per Seat Month | - | - | $20.00 | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Gemini 3.1 Pro gemini-3.1-pro Compute ROI → | Llm Api | Multimodal | Frontier | 2,000,000 tokens | - | - | Direct | Needs Verification | 2M context (largest in industry). Price unconfirmed; AICostCompass table shows $3.50/$14 — reconcile with Google. | No current price | - | - | - | USD | - | Missing | Missing | Missing | |
| Gemini 3 Pro gemini-3-pro Compute ROI → | Llm Api | Multimodal | Frontier | 1,000,000 tokens | - | - | Direct | Priced | Long-context tier (>200K tokens): $4 input / $18 output per 1M. | Per 1m Tokens | $2.00 | $12.00 | - | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source | |
| Gemini 3.5 Flash gemini-3.5-flash Compute ROI → | Llm Api | Multimodal | Fast | Unknown | - | - | Direct | Needs Verification | Newer Flash line; pricing to confirm. | No current price | - | - | - | USD | - | Missing | Missing | Missing | |
| Gemini 3 Flash gemini-3-flash Compute ROI → | Llm Api | Multimodal | Budget | Unknown | - | - | Direct | Priced | Cheap workhorse. | Per 1m Tokens | $0.50 | $3.00 | - | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source | |
| Gemini 2.5 Flash-Lite gemini-2.5-flash-lite Compute ROI → | Llm Api | Multimodal | Budget | Unknown | - | - | Direct | Needs Verification | Among cheapest production-grade models; output price to confirm. | No current price | - | - | - | USD | - | Missing | Missing | Missing | |
| Google Antigravity | Antigravity Pro antigravity-pro Compute ROI → | Coding Tool | Code | Balanced | Unknown | - | - | Direct | Priced | Standard paid tier. | Per Seat Month | - | - | $20.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Google Antigravity | Antigravity Ultra antigravity-ultra Compute ROI → | Coding Tool | Code | Frontier | Unknown | - | - | Direct | Priced | ~5x Pro quotas; split out from the old single Ultra tier at Google I/O 2026. | Per Seat Month | - | - | $100.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Google Antigravity | Antigravity Ultra Max antigravity-ultra-max Compute ROI → | Coding Tool | Code | Frontier | Unknown | - | - | Direct | Priced | ~20x Pro quotas; cut from $249.99 to $200 at Google I/O 2026. | Per Seat Month | - | - | $200.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Groq | Llama 3.3 70B Versatile llama-3.3-70b-versatile Compute ROI → | Llm Api | Text | Balanced | Unknown | - | - | Direct | Priced | Flagship on Groq LPU inference; open-weight Llama model. | Per 1m Tokens | $0.59 | $0.79 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Groq | GPT-OSS 120B gpt-oss-120b Compute ROI → | Llm Api | Text | Budget | Unknown | - | - | Direct | Priced | OpenAI open-weight model on Groq LPU. | Per 1m Tokens | $0.15 | $0.60 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Groq | Llama 3.1 8B Instant llama-3.1-8b-instant Compute ROI → | Llm Api | Text | Budget | Unknown | - | - | Direct | Priced | Cheapest Groq-hosted model. | Per 1m Tokens | $0.05 | $0.08 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| JetBrains AI Assistant | AI Pro ai-pro Compute ROI → | Coding Tool | Code | Balanced | Unknown | - | - | Direct | Priced | Individual tier; unlimited credits, Junie agent, multi-model switching. | Per Seat Month | - | - | $10.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| JetBrains AI Assistant | AI Ultimate ai-ultimate Compute ROI → | Coding Tool | Code | Frontier | Unknown | - | - | Direct | Priced | Individual tier; priority model access, advanced agent capabilities. | Per Seat Month | - | - | $30.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Kiro (AWS) | Kiro Pro kiro-pro Compute ROI → | Coding Tool | Code | Balanced | Unknown | - | - | Direct | Priced | Credit-based; overages at $0.04/credit. | Per Seat Month | - | - | $20.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Kiro (AWS) | Kiro Pro+ kiro-pro-plus Compute ROI → | Coding Tool | Code | Balanced | Unknown | - | - | Direct | Priced | Credit-based; overages at $0.04/credit. | Per Seat Month | - | - | $40.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Kiro (AWS) | Kiro Pro Max kiro-pro-max Compute ROI → | Coding Tool | Code | Frontier | Unknown | - | - | Direct | Priced | Credit-based; overages at $0.04/credit. | Per Seat Month | - | - | $100.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Kiro (AWS) | Kiro Power kiro-power Compute ROI → | Coding Tool | Code | Frontier | Unknown | - | - | Direct | Priced | Top tier; credit-based, overages at $0.04/credit. | Per Seat Month | - | - | $200.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Meta | Llama 3.3 70B llama-3.3-70b Compute ROI → | Cloud Hosting | Text | Balanced | 128,000 tokens | - | - | Meta | Needs Verification | Open weights; price host-dependent (~$0.72 Bedrock, ~$0.59-0.79 Azure). Blended figure. | No current price | - | - | - | USD | - | Missing | Missing | Missing |
| Meta | Llama 4 (Scout / Maverick) llama-4-maverick Compute ROI → | Cloud Hosting | Multimodal | Balanced | Unknown | - | - | Meta | Needs Verification | Newer Llama 4 family; priced per host. | No current price | - | - | - | USD | - | Missing | Missing | Missing |
| Mistral | Mistral Large 3 large-3 Compute ROI → | Llm Api | Text | Frontier | Unknown | - | - | Direct | Needs Verification | Cheapest flagship-class model; output price to confirm. | No current price | - | - | - | USD | - | Missing | Missing | Missing |
| Mistral | Mistral Large 2 large-2 Compute ROI → | Llm Api | Text | Balanced | Unknown | - | - | Direct | Priced | Prior flagship. | Per 1m Tokens | $2.00 | $6.00 | - | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Mistral | Mistral Small small Compute ROI → | Llm Api | Text | Budget | Unknown | - | - | Direct | Priced | Small tier. | Per 1m Tokens | $0.20 | $0.60 | - | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Mistral | Ministral 3B ministral-3b Compute ROI → | Llm Api | Text | Budget | Unknown | - | - | Direct | Priced | Edge/cheap model. | Per 1m Tokens | $0.04 | $0.04 | - | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| Moonshot AI (Kimi) | Kimi K2.6 kimi-k2.6 Compute ROI → | Llm Api | Text | Balanced | Unknown | - | Cache input: $0.16 / 1M | Direct | Priced | Flagship Kimi model; OpenAI-compatible API. | Per 1m Tokens | $0.95 | $4.00 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Moonshot AI (Kimi) | Kimi K2.5 kimi-k2.5 Compute ROI → | Llm Api | Text | Budget | Unknown | - | Cache input: $0.10 / 1M | Direct | Priced | Cheaper prior-gen Kimi model. | Per 1m Tokens | $0.60 | $3.00 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| OpenAI | GPT-5.6 Sol gpt-5.6-sol Compute ROI → | Llm Api | Multimodal | Frontier | 1,050,000 tokens | 128,000 tokens | Cache input: $0.50 / 1M | Direct | Priced | GA 2026-07-09. Cache read -90%; explicit cache writes 1.25x input. Output price conflicts across sources ($6 vs $30) — verify. | Per 1m Tokens | $5.00 | $30.00 | - | USD | Cache 90% | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| OpenAI | GPT-5.6 Terra gpt-5.6-terra Compute ROI → | Llm Api | Multimodal | Balanced | 1,050,000 tokens | 128,000 tokens | - | Direct | Priced | Balanced production tier of the GPT-5.6 family. | Per 1m Tokens | $2.50 | $15.00 | - | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| OpenAI | GPT-5.6 Luna gpt-5.6-luna Compute ROI → | Llm Api | Multimodal | Budget | 1,050,000 tokens | 128,000 tokens | - | Direct | Priced | Cost/volume tier of the GPT-5.6 family. | Per 1m Tokens | $1.00 | $6.00 | - | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| OpenAI | GPT-5.5 gpt-5.5 Compute ROI → | Llm Api | Multimodal | Frontier | 400,000 tokens | - | Cache input: $0.50 / 1M | Direct | Priced | Previous flagship; context window to verify. | Per 1m Tokens | $5.00 | $30.00 | - | USD | Cache 90% | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| OpenAI | GPT-5.4 gpt-5.4 Compute ROI → | Llm Api | Multimodal | Balanced | Unknown | - | - | Direct | Needs Verification | Still offered; pricing to confirm. | No current price | - | - | - | USD | - | Missing | Missing | Missing |
| OpenAI | GPT-5.3 Codex gpt-5.3-codex Compute ROI → | Llm Api | Code | Reasoning | Unknown | - | Cache input: $0.175 / 1M | Direct | Priced | Coding-optimised API model behind the Codex CLI/IDE extension. Priority-tier pricing is 2x standard ($3.50 in / $28.00 out, $0.35 cached). | Per 1m Tokens | $1.75 | $14.00 | - | USD | - | 7/15/2026 | 7/15/2026, 12:00:00 AM Refreshing | Source |
| OpenAI Codex | Codex Go codex-go Compute ROI → | Coding Tool | Code | Budget | Unknown | - | - | Direct | Priced | Lightweight coding tasks tier. | Per Seat Month | - | - | $8.00 | USD | - | 7/15/2026 | 7/15/2026, 12:00:00 AM Refreshing | Source |
| OpenAI Codex | Codex Plus codex-plus Compute ROI → | Coding Tool | Code | Balanced | Unknown | - | - | Direct | Priced | Shares the $20/mo ChatGPT Plus seat; includes GPT-5.6 models plus cloud/CLI/IDE Codex access. | Per Seat Month | - | - | $20.00 | USD | - | 7/15/2026 | 7/15/2026, 12:00:00 AM Refreshing | Source |
| OpenAI Codex | Codex Pro codex-pro Compute ROI → | Coding Tool | Code | Frontier | Unknown | - | - | Direct | Priced | From $100/mo; 5x-20x higher rate limits than Plus, access to GPT-5.3-Codex-Spark. | Per Seat Month | - | - | $100.00 | USD | - | 7/15/2026 | 7/15/2026, 12:00:00 AM Refreshing | Source |
| Perplexity | Sonar sonar Compute ROI → | Llm Api | Text | Balanced | Unknown | - | - | Direct | Priced | Plus a per-request search fee ($5-12/1K requests depending on search context size) not reflected in the token price. | Per 1m Tokens | $1.00 | $1.00 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Perplexity | Sonar Pro sonar-pro Compute ROI → | Llm Api | Text | Frontier | Unknown | - | - | Direct | Priced | Plus a per-request search fee ($6-14/1K requests depending on search context size) not reflected in the token price. | Per 1m Tokens | $3.00 | $15.00 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Perplexity | Sonar Reasoning Pro sonar-reasoning-pro Compute ROI → | Llm Api | Text | Reasoning | Unknown | - | - | Direct | Priced | Plus a per-request search fee ($6-14/1K requests depending on search context size) not reflected in the token price. | Per 1m Tokens | $2.00 | $8.00 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Replit Agent | Replit Core replit-core Compute ROI → | Coding Tool | Code | Balanced | Unknown | - | - | Direct | Priced | Includes $20/mo of usage credits, up to 5 collaborators; effort-based Agent credit pricing on top. | Per Seat Month | - | - | $25.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Replit Agent | Replit Pro replit-pro Compute ROI → | Coding Tool | Code | Frontier | Unknown | - | - | Direct | Priced | Flat team price for up to 15 builders with pooled/tiered credits; replaces the old Teams plan. | Per Seat Month | - | - | $100.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Sourcegraph Cody | Cody Enterprise cody-enterprise Compute ROI → | Coding Tool | Code | Frontier | Unknown | - | - | Direct | Priced | Enterprise-only since mid-2025 (Free/Pro discontinued); this is a reported reference price, actual pricing is negotiated per customer. | Per Seat Month | - | - | $59.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Tabnine | Code Assistant code-assistant Compute ROI → | Coding Tool | Code | Balanced | Unknown | - | - | Direct | Priced | Requires annual billing; free Basic/Dev tiers discontinued in 2025. | Per Seat Month | - | - | $39.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Tabnine | Agentic Platform agentic-platform Compute ROI → | Coding Tool | Code | Frontier | Unknown | - | - | Direct | Priced | Adds agentic workflows and the Tabnine Context Engine. | Per Seat Month | - | - | $59.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Together AI | DeepSeek V4 Pro deepseek-v4-pro Compute ROI → | Llm Api | Text | Balanced | Unknown | - | - | Direct | Priced | Hosted open-weight model. | Per 1m Tokens | $1.74 | $3.48 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Together AI | Qwen3.7 Plus qwen3.7-plus Compute ROI → | Llm Api | Text | Balanced | Unknown | - | - | Direct | Priced | Together-hosted rate for this open-weight model; differs from Alibaba’s own $0.40/$1.60 direct pricing. | Per 1m Tokens | $0.32 | $1.28 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Together AI | Kimi K2.7 Code kimi-k2.7-code Compute ROI → | Llm Api | Code | Reasoning | Unknown | - | - | Direct | Priced | Coding-optimised Kimi variant, hosted serverless. | Per 1m Tokens | $0.95 | $4.00 | - | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Windsurf (Cognition) | Windsurf Pro windsurf-pro Compute ROI → | Coding Tool | Code | Balanced | Unknown | - | - | Direct | Priced | Overhauled to daily/weekly quotas (from credits) on 2026-03-19; includes SWE-1.5 model + cloud sessions. | Per Seat Month | - | - | $20.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| Windsurf (Cognition) | Windsurf Max windsurf-max Compute ROI → | Coding Tool | Code | Frontier | Unknown | - | - | Direct | Priced | Heavy-usage tier above Pro. | Per Seat Month | - | - | $200.00 | USD | - | 7/16/2026 | 7/16/2026, 12:00:00 AM Refreshing | Source |
| xAI | Grok 4.5 grok-4.5 Compute ROI → | Llm Api | Multimodal | Frontier | 500,000 tokens | - | Cache input: $0.50 / 1M | Direct | Priced | Flagship since 2026-07-08. Cache -75%. | Per 1m Tokens | $2.00 | $6.00 | - | USD | Cache 75% | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| xAI | Grok 4.3 grok-4.3 Compute ROI → | Llm Api | Text | Balanced | Unknown | - | - | Direct | Priced | Cheaper long-context option. | Per 1m Tokens | $1.25 | $2.50 | - | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| xAI | Grok 4.1 Fast grok-4.1-fast Compute ROI → | Llm Api | Text | Budget | 2,000,000 tokens | - | - | Direct | Priced | Budget workhorse; 2M context. | Per 1m Tokens | $0.20 | $0.50 | - | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| xAI | Grok 4 grok-4 Compute ROI → | Llm Api | Multimodal | Frontier | 256,000 tokens | - | - | Direct | Priced | Original flagship. | Per 1m Tokens | $3.00 | $15.00 | - | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
| xAI | Grok Code Fast 1 grok-code-fast-1 Compute ROI → | Llm Api | Code | Budget | 256,000 tokens | - | - | Direct | Priced | Coding-optimised; figures per AICostCompass data — confirm on x.ai. | Per 1m Tokens | $0.20 | $1.50 | - | USD | - | 7/14/2026 | 7/14/2026, 12:00:00 AM Refreshing | Source |
How to use this comparison
Start with the benchmark workload profile above the chart — it applies a realistic monthly token or usage mix so models are compared on cost for actual work, not on a single unit price. The chart plots cost against quality tier, so the models worth a closer look sit toward the lower-right: strong quality at low monthly cost. Click any point or table row to open that model's full price history and source records, or jump straight to an ROI estimate for it.
The table below the chart is the same data in a sortable, filterable form. Use the provider, modality, and quality-tier filters to narrow a long list before reading unit prices, context window, and discount columns. If you already know your own monthly usage rather than the default benchmark, use the comparison tool — it accepts your numbers directly instead of the benchmark assumption.
Where this pricing data comes from
Every price shown here is read directly from a provider's own pricing page or API documentation, never estimated or inferred. Each ingestion run stamps the row with the exact time it was checked, and the table's Source column links back to the page the number came from so you can verify it independently. Price history is never edited in place — when a rate changes, the previous price is kept as a historical record and the new one is added separately — and any change large enough to look like a possible data error is held for manual confirmation before it's published, rather than shown unverified.
Pricing comparison FAQ
How often is the pricing data updated?
The catalog refreshes hourly. Each row in the table shows the exact "last verified" timestamp alongside a link to the provider source it was read from, so you can confirm a price yourself rather than trusting a stale snapshot.
Do I need an account to compare prices?
No. Browsing, filtering, reading the chart, copying share links, and using the ROI calculator are all account-free.
What does "quality tier" mean on the chart and table?
It's a rough capability class — frontier, balanced, or budget — used to group models with similar strengths, not a numeric benchmark score. Use it to avoid comparing a narrow classifier against a frontier reasoning model as if they were substitutes.
Why do some models show no current price?
When a provider changes a rate by more than a review threshold, the new price is held for manual confirmation instead of being published automatically, so a model can briefly show no current price rather than an unverified one.
Is the price shown here what I'll actually be billed?
Treat it as an informational estimate, not an invoice. Provider pricing pages are the source of truth — this table exists to help you compare quickly, not to replace checking directly before you commit spend.