AI models
Every model in the CertuAI chat, with how it performed on real requests this week: latency, time to first token, throughput and uptime. Price per million tokens and context window next to it.
| Model | Tokens (7d) | Latency p50 | First token p50 | Throughput | Uptime | Input / 1M | Output / 1M | Context |
|---|---|---|---|---|---|---|---|---|
| Claude Fable 5Anthropic | — | — | — | — | — | $10.00 | $50.00 | 1M |
| Claude Fable 5.1Anthropic | — | — | — | — | — | $10.00 | $50.00 | 1M |
| Claude Haiku 4.5Anthropic | — | — | — | — | — | $1.00 | $5.00 | 200K |
| Claude Mythos 5Anthropic | — | — | — | — | — | $10.00 | $50.00 | 1M |
| Claude Mythos 5.1Anthropic | — | — | — | — | — | $10.00 | $50.00 | 1M |
| Claude Opus 4.5Anthropic | — | — | — | — | — | $5.00 | $25.00 | 200K |
| Claude Opus 4.6Anthropic | — | — | — | — | — | $5.00 | $25.00 | 1M |
| Claude Opus 4.7Anthropic | — | — | — | — | — | $5.00 | $25.00 | 1M |
| Claude Opus 4.8Anthropic | — | — | — | — | — | $5.00 | $25.00 | 1M |
| Claude Opus 5Anthropic | — | — | — | — | — | $5.00 | $25.00 | 1M |
| Claude Opus 5.5Anthropic | — | — | — | — | — | $4.00 | $20.00 | 1M |
| Claude Sonnet 4.5Anthropic | — | — | — | — | — | $3.00 | $15.00 | 200K |
| Claude Sonnet 4.6Anthropic | — | — | — | — | — | $3.00 | $15.00 | 1M |
| Claude Sonnet 5Anthropic | — | — | — | — | — | $3.00 | $15.00 | 1M |
| DeepSeek R1 0528DeepSeek | — | — | — | — | — | $0.50 | $2.15 | 163.8K |
| DeepSeek V3DeepSeek | — | — | — | — | — | $0.32 | $0.89 | 163.8K |
| DeepSeek V3.1DeepSeek | — | — | — | — | — | $0.25 | $0.95 | 163.8K |
| DeepSeek V3.2DeepSeek | — | — | — | — | — | $0.26 | $0.38 | 163.8K |
| DeepSeek V4 FlashDeepSeek | — | — | — | — | — | $0.090 | $0.18 | 1M |
| DeepSeek V4 Flash (free)FreeDeepSeek | — | — | — | — | — | $0.090 | $0.18 | 32.8K |
| DeepSeek V4 Flash 0731DeepSeek | — | — | — | — | — | $0.060 | $0.18 | 1M |
| DeepSeek V4 Flash Vision ExpDeepSeek | — | — | — | — | — | $0.44 | $1.32 | 1M |
| DeepSeek V4 ProDeepSeek | — | — | — | — | — | $1.30 | $2.60 | 1M |
| DeepSeek V4 Pro 0813DeepSeek | — | — | — | — | — | $1.30 | $2.60 | 1M |
| DeepSeek V4.1 FlashDeepSeek | — | — | — | — | — | $0.20 | $0.60 | 1M |
| Free (auto)FreeCertu | — | — | — | — | — | $0.000 | $0.000 | 32.8K |
| Gemini 2.5 FlashGoogle | — | — | — | — | — | $0.30 | $2.50 | 1M |
| Gemini 2.5 Flash LiteGoogle | — | — | — | — | — | $0.10 | $0.40 | 1M |
| Gemini 2.5 ProGoogle | — | — | — | — | — | $1.25 | $10.00 | 1M |
| Gemini 3 Flash PreviewGoogle | — | — | — | — | — | $0.50 | $3.00 | 1M |
| Gemini 3.1 Flash LiteGoogle | — | — | — | — | — | $0.25 | $1.50 | 1M |
| Gemini 3.1 Pro PreviewGoogle | — | — | — | — | — | $2.00 | $12.00 | 1M |
| Gemini 3.5 FlashGoogle | — | — | — | — | — | $1.50 | $9.00 | 1M |
| Gemini 3.5 Flash LiteGoogle | — | — | — | — | — | $0.30 | $2.50 | 1M |
| Gemini 3.6 FlashGoogle | — | — | — | — | — | $0.75 | $3.75 | 1M |
| Gemini 3.7 FlashGoogle | — | — | — | — | — | $0.75 | $3.75 | 1M |
| Gemini 3.8 FlashGoogle | — | — | — | — | — | $0.75 | $3.75 | 1M |
| gemma 3 12b itGoogle | — | — | — | — | — | $0.050 | $0.15 | 131.1K |
| gemma 3 27b itGoogle | — | — | — | — | — | $0.080 | $0.16 | 131.1K |
| gemma 3 4b itGoogle | — | — | — | — | — | $0.050 | $0.10 | 131.1K |
| gemma 4 26B A4B itGoogle | — | — | — | — | — | $0.070 | $0.34 | 262.1K |
| Gemma 4 31BGoogle | — | — | — | — | — | $0.20 | $0.40 | 262.1K |
| gemma 4 31B it turboGoogle | — | — | — | — | — | $0.090 | $0.34 | 262.1K |
| gemma 4 31B it UltraGoogle | — | — | — | — | — | $0.27 | $0.76 | 131.1K |
| GLM 4.6Z.ai | — | — | — | — | — | $0.50 | $2.00 | 202.8K |
| GLM 4.7Z.ai | — | — | — | — | — | $0.40 | $1.75 | 202.8K |
| GLM 5.2Z.ai | — | — | — | — | — | $0.75 | $2.40 | 1M |
| gpt oss 120bOpenAI | — | — | — | — | — | $0.037 | $0.17 | 131.1K |
| gpt oss 120b (free)FreeOpenAI | — | — | — | — | — | $0.037 | $0.17 | 32.8K |
| gpt oss 120b TurboOpenAI | — | — | — | — | — | $0.15 | $0.60 | 131.1K |
| gpt oss 120b UltraOpenAI | — | — | — | — | — | $0.20 | $0.95 | 131.1K |
| gpt oss 20bOpenAI | — | — | — | — | — | $0.030 | $0.14 | 131.1K |
| GPT-4-turbo-2024-04-09OpenAI | — | — | — | — | — | $10.00 | $30.00 | 128K |
| GPT-4.1OpenAI | — | — | — | — | — | $2.00 | $8.00 | 1M |
| GPT-4.1-miniOpenAI | — | — | — | — | — | $0.40 | $1.60 | 1M |
| GPT-4.1-nanoOpenAI | — | — | — | — | — | $0.10 | $0.40 | 1M |
| GPT-4oOpenAI | — | — | — | — | — | $2.50 | $10.00 | 128K |
| GPT-4o-miniOpenAI | — | — | — | — | — | $0.15 | $0.60 | 128K |
| GPT-5OpenAI | — | — | — | — | — | $1.25 | $10.00 | 400K |
| GPT-5-miniOpenAI | — | — | — | — | — | $0.25 | $2.00 | 400K |
| GPT-5-nanoOpenAI | — | — | — | — | — | $0.050 | $0.40 | 400K |
| GPT-5.1OpenAI | — | — | — | — | — | $1.25 | $10.00 | 400K |
| GPT-5.2OpenAI | — | — | — | — | — | $1.75 | $14.00 | 400K |
| GPT-5.4OpenAI | — | — | — | — | — | $2.50 | $15.00 | 1.1M |
| GPT-5.4-miniOpenAI | — | — | — | — | — | $0.75 | $4.50 | 400K |
| GPT-5.4-nanoOpenAI | — | — | — | — | — | $0.20 | $1.25 | 400K |
| GPT-5.5OpenAI | — | — | — | — | — | $5.00 | $30.00 | 1.1M |
| GPT-5.6-lunaOpenAI | — | — | — | — | — | $0.20 | $1.20 | 1.1M |
| GPT-5.6-solOpenAI | — | — | — | — | — | $4.00 | $20.00 | 1.1M |
| GPT-5.6-terraOpenAI | — | — | — | — | — | $2.00 | $12.00 | 1.1M |
| GPT-6-astraOpenAI | — | — | — | — | — | $10.00 | $50.00 | 1.1M |
| GPT-6-lunaOpenAI | — | — | — | — | — | $0.10 | $0.50 | 1.1M |
| GPT-6-solOpenAI | — | — | — | — | — | $2.00 | $10.00 | 1.1M |
| granite 4.2 8bIBM | — | — | — | — | — | $0.060 | $0.25 | 131.1K |
| Hermes 3 Llama 3.1 405BNous Research | — | — | — | — | — | $1.00 | $1.00 | 131.1K |
| Hermes 3 Llama 3.1 70BNous Research | — | — | — | — | — | $0.70 | $0.70 | 131.1K |
| Hy4 previewTencent | — | — | — | — | — | $0.83 | $2.50 | 1M |
| InklingThinking Machines | — | — | — | — | — | $0.95 | $4.05 | 524.3K |
| Inkling SmallThinking Machines | — | — | — | — | — | $0.45 | $1.20 | 524.3K |
| Kimi K2.6Moonshot | — | — | — | — | — | $0.75 | $3.50 | 262.1K |
| Kimi K3Moonshot | — | — | — | — | — | $2.85 | $14.25 | 1M |
| L3.1 70B Euryale v2.2Sao10K | — | — | — | — | — | $0.85 | $0.85 | 131.1K |
| Ling 3.0 flashinclusionAI | — | — | — | — | — | $0.060 | $0.18 | 131.1K |
| Ling 3.0 flash VLinclusionAI | — | — | — | — | — | $0.060 | $0.18 | 131.1K |
| Ling 3.1 flashinclusionAI | — | — | — | — | — | $0.30 | $0.90 | 262.1K |
| Llama 3.3 70BMeta | — | — | — | — | — | $0.10 | $0.32 | 131.1K |
| Llama 3.3 70B (free)FreeMeta | — | — | — | — | — | $0.10 | $0.32 | 32.8K |
| Llama 4 ScoutMeta | — | — | — | — | — | $0.10 | $0.30 | 327.7K |
| Meta Llama 3.1 70B Instruct TurboMeta | — | — | — | — | — | $0.40 | $0.40 | 131.1K |
| Meta Llama 3.1 8B Instruct TurboMeta | — | — | — | — | — | $0.020 | $0.040 | 131.1K |
| MiMo V2.6 FlashXiaomi | — | — | — | — | — | $0.14 | $0.28 | 1M |
| MiMo V2.6 ProXiaomi | — | — | — | — | — | $0.43 | $0.87 | 1M |
| MiniMax M3MiniMax | — | — | — | — | — | $0.28 | $1.10 | 524.3K |
| Mistral Nemo Instruct 2407Mistral | — | — | — | — | — | $0.029 | $0.030 | 131.1K |
| Mistral Small 24B Instruct 2501Mistral | — | — | — | — | — | $0.050 | $0.080 | 32.8K |
| Mistral Small 3.2Mistral | — | — | — | — | — | $0.075 | $0.20 | 128K |
| Mistral Small 3.2 (free)FreeMistral | — | — | — | — | — | $0.075 | $0.20 | 32.8K |
| Muse Glimmer 30BMeta Models | — | — | — | — | — | $0.30 | $1.20 | 131.1K |
| Nemotron 3 UltraNVIDIA | — | — | — | — | — | $0.50 | $2.20 | 262.1K |
| NVIDIA Nemotron 3 Super 120B A12BNVIDIA | — | — | — | — | — | $0.085 | $0.40 | 262.1K |
| NVIDIA Nemotron 3.5 LightningNVIDIA | — | — | — | — | — | $0.060 | $0.16 | 262.1K |
| o1OpenAI | — | — | — | — | — | $15.00 | $60.00 | 200K |
| o3OpenAI | — | — | — | — | — | $2.00 | $8.00 | 200K |
| o3-miniOpenAI | — | — | — | — | — | $1.10 | $4.40 | 200K |
| o4-miniOpenAI | — | — | — | — | — | $1.10 | $4.40 | 200K |
| Qwen 3.8 FlashAlibaba | — | — | — | — | — | $0.11 | $0.38 | 1M |
| Qwen 3.8 Flash (free)FreeAlibaba | — | — | — | — | — | $0.11 | $0.38 | 32.8K |
| Qwen 3.8 MaxAlibaba | — | — | — | — | — | $1.65 | $4.95 | 256K |
| Qwen2.5 72B InstructAlibaba | — | — | — | — | — | $0.36 | $0.40 | 32.8K |
| Qwen3 14BAlibaba | — | — | — | — | — | $0.12 | $0.24 | 41K |
| Qwen3 235B A22B Instruct 2507Alibaba | — | — | — | — | — | $0.090 | $0.55 | 262.1K |
| Qwen3 30B A3BAlibaba | — | — | — | — | — | $0.12 | $0.50 | 41K |
| Qwen3 32BAlibaba | — | — | — | — | — | $0.080 | $0.28 | 41K |
| Qwen3 Coder 480B A35B Instruct TurboAlibaba | — | — | — | — | — | $0.30 | $1.00 | 262.1K |
| Qwen3 Next 80B A3B InstructAlibaba | — | — | — | — | — | $0.090 | $1.10 | 262.1K |
| Qwen3 VL 235B A22B InstructAlibaba | — | — | — | — | — | $0.20 | $0.88 | 262.1K |
| Qwen3 VL 30B A3B InstructAlibaba | — | — | — | — | — | $0.15 | $0.60 | 262.1K |
| Qwen3.5 27BAlibaba | — | — | — | — | — | $0.26 | $2.60 | 262.1K |
| Qwen3.5 35B A3BAlibaba | — | — | — | — | — | $0.14 | $1.00 | 262.1K |
| Qwen3.5 397B A17BAlibaba | — | — | — | — | — | $0.45 | $3.00 | 262.1K |
| Qwen3.5 9BAlibaba | — | — | — | — | — | $0.10 | $0.15 | 262.1K |
| Qwen3.6 27BAlibaba | — | — | — | — | — | $0.32 | $3.20 | 262.1K |
| Qwen3.6 35B A3BAlibaba | — | — | — | — | — | $0.10 | $0.95 | 262.1K |
| Qwen3.7 MaxAlibaba | — | — | — | — | — | $2.50 | $7.50 | 256K |
| Qwen3.8 27BAlibaba | — | — | — | — | — | $0.20 | $2.50 | 262.1K |
| Seed 1.8ByteDance | — | — | — | — | — | $0.25 | $2.00 | 256K |
| Seed 2.0 codeByteDance | — | — | — | — | — | $0.50 | $3.00 | 256K |
| Seed 2.0 miniByteDance | — | — | — | — | — | $0.10 | $0.40 | 256K |
| Seed 2.0 proByteDance | — | — | — | — | — | $0.50 | $3.00 | 256K |
| Step 5 PreviewStepFun | — | — | — | — | — | $1.00 | $2.70 | 1M |
No model matches these filters.
How these numbers are measured
Every request the chat and the API send to a model is counted the moment it finishes: tokens in and out, total latency (from the request to the last byte), time to first token (streamed answers) and throughput (output tokens per second, counted after the first token, only on answers with at least 20 tokens). Medians are estimated from a latency histogram, so they are close but not exact.
Uptime is 1 minus the share of requests the provider failed (timeouts, rate limits, server errors and network failures). A request rejected for being malformed does not count against the provider. Uptime is shown only after 50 requests in the period, and latency only after 10 answers, so a model with three lucky requests never tops the list.
Days are in UTC. Today counts from midnight UTC; this week is the last 7 days including today. Synthetic checks, when enabled, count for uptime and speed but never for popularity.