{"asOf":"2026-06-28","methodology":"A model is listed as a good free option only if it is callable at $0 on at least one hosted provider and clears the quality gate: an Artificial Analysis Intelligence Index of at least 20 of 100. That index (v4.1) is the one benchmark scored for every model here; it blends nine independent evaluations including coding, so a single number ranks the table consistently. Benchmark scores are scraped from the public Artificial Analysis leaderboard; free-tier terms are verified against each provider source and dated.","qualityGate":{"metric":"Artificial Analysis Intelligence Index","min":20,"note":"The index is demanding: the best model scores about 60 and the strongest open-weight model about 51, so the 0-100 scale runs low. A free model scoring at least 20 is genuinely capable; smaller and older free models (Gemma 4 E-series, tiny Qwen, Llama 3.x) fall below and are not listed."},"benchmarkSource":{"name":"Artificial Analysis","url":"https://artificialanalysis.ai/leaderboards/models","metric":"Artificial Analysis Intelligence Index","version":"v4.1","note":"The Artificial Analysis Intelligence Index (v4.1) is a single 0-100 score blending nine independent evaluations (GPQA Diamond, Terminal-Bench, SciCode, AA-LCR, Humanity's Last Exam, GDPval, t3-Banking, CritPt, AA-Omniscience), so one number reflects reasoning, knowledge and coding together. It is the one benchmark that scores every model on this page. Read from the public leaderboard with Playwright on the date below; this site does not run the evaluations.","scoresVerified":"2026-06-28"},"models":[{"model":"Gemini 3.5 Flash","maker":"Google","aaIndex":50,"terminalBench":79,"sciCode":53,"contextLength":1048576,"openWeight":false,"hosts":[{"key":"google-ai-studio","name":"Google AI Studio","accessUrl":"https://aistudio.google.com/"}],"note":"The top free model on this page by the AA Index. Closed weights, free on Google AI Studio's Flash tier (Gemini Pro models left the free tier on 2026-04-01). Score is the standard (high) reasoning configuration."},{"model":"MiMo V2.5","maker":"Xiaomi","aaIndex":40,"terminalBench":null,"sciCode":43,"contextLength":1048576,"openWeight":false,"hosts":[{"key":"opencode-zen","name":"OpenCode Zen","accessUrl":"https://opencode.ai/docs/zen/"}],"note":"Xiaomi's omnimodal model, free on OpenCode Zen (data may be used to improve the model). The paid MiMo V2.5 Pro sibling scores 42 on the same index."},{"model":"Nemotron 3 Ultra 550B","maker":"NVIDIA","aaIndex":38,"terminalBench":54,"sciCode":40,"contextLength":1000000,"openWeight":true,"hosts":[{"key":"openrouter","name":"OpenRouter","accessUrl":"https://openrouter.ai/models?max_price=0"},{"key":"opencode-zen","name":"OpenCode Zen","accessUrl":"https://opencode.ai/docs/zen/"},{"key":"nvidia-nim","name":"NVIDIA NIM (build.nvidia.com)","accessUrl":"https://build.nvidia.com/"}],"note":"Open MoE (55B active of 550B) and the strongest open-weight free model here. Free on OpenRouter, on OpenCode Zen (trial), and on NVIDIA's own endpoints."},{"model":"DeepSeek V4 Flash","maker":"DeepSeek","aaIndex":29,"terminalBench":null,"sciCode":37,"contextLength":1048576,"openWeight":true,"hosts":[{"key":"opencode-zen","name":"OpenCode Zen","accessUrl":"https://opencode.ai/docs/zen/"}],"note":"Efficiency-optimized open MoE (13B active of 284B) with a 1M-token context. Free on OpenCode Zen (data may be used to improve the model). The paid V4 Pro sibling scores 44; the Max-effort config of Flash scores 40."},{"model":"Gemma 4 31B","maker":"Google","aaIndex":29,"terminalBench":43,"sciCode":43,"contextLength":262144,"openWeight":true,"hosts":[{"key":"openrouter","name":"OpenRouter","accessUrl":"https://openrouter.ai/models?max_price=0"}],"note":"Google's open-weight Gemma 4, free on OpenRouter. Strong for its size and small enough to also self-host."},{"model":"Nemotron 3 Super 120B","maker":"NVIDIA","aaIndex":25,"terminalBench":39,"sciCode":36,"contextLength":1000000,"openWeight":true,"hosts":[{"key":"openrouter","name":"OpenRouter","accessUrl":"https://openrouter.ai/models?max_price=0"},{"key":"nvidia-nim","name":"NVIDIA NIM (build.nvidia.com)","accessUrl":"https://build.nvidia.com/"}],"note":"Open MoE (12B active of 120B), the smaller free Nemotron. Free on OpenRouter and NVIDIA NIM."},{"model":"Gemini 3.1 Flash-Lite","maker":"Google","aaIndex":25,"terminalBench":31,"sciCode":42,"contextLength":1048576,"openWeight":false,"hosts":[{"key":"google-ai-studio","name":"Google AI Studio","accessUrl":"https://aistudio.google.com/"}],"note":"The fastest, highest-quota free Gemini tier on Google AI Studio (500 requests/day). Closed weights."},{"model":"gpt-oss-120b","maker":"OpenAI","aaIndex":24,"terminalBench":26,"sciCode":39,"contextLength":131072,"openWeight":true,"hosts":[{"key":"groq","name":"Groq","accessUrl":"https://console.groq.com/"},{"key":"cerebras","name":"Cerebras","accessUrl":"https://cloud.cerebras.ai/"},{"key":"cloudflare","name":"Cloudflare Workers AI","accessUrl":"https://developers.cloudflare.com/workers-ai/"},{"key":"openrouter","name":"OpenRouter","accessUrl":"https://openrouter.ai/models?max_price=0"}],"note":"Apache-2.0 open MoE (5.1B active of 117B), runs on a single H100. Score is the high-reasoning-effort config; the low-effort config scores 18. The most widely free-hosted model here, and the fastest via Groq and Cerebras."},{"model":"Qwen3 Next 80B A3B","maker":"Qwen (Alibaba)","aaIndex":20,"terminalBench":null,"sciCode":39,"contextLength":262144,"openWeight":true,"hosts":[{"key":"openrouter","name":"OpenRouter","accessUrl":"https://openrouter.ai/models?max_price=0"}],"note":"Open MoE (3B active of 80B) with a 262K context, free on OpenRouter as the Instruct variant. Efficient and cheap to self-host."}],"hosts":[{"key":"google-ai-studio","name":"Google AI Studio","freeSummary":"Gemini Flash and Flash-Lite free (Pro tiers left the free tier on 2026-04-01); roughly 5-15 req/min and 20-1,500 req/day depending on model","requiresSignup":true,"phoneVerify":false,"dataUsedForTraining":true,"speedNote":"","accessUrl":"https://aistudio.google.com/","sourceUrl":"https://ai.google.dev/gemini-api/docs/rate-limits","verifiedDate":"2026-06-28","note":"Free-tier inputs may be used to improve Google products (outside the UK, CH, EEA, and EU). Gemini 2.5/3.x Pro are no longer free as of 2026-04-01."},{"key":"openrouter","name":"OpenRouter","freeSummary":"Models tagged :free at 20 req/min, 50 req/day (1,000/day after a one-time $10 credit purchase)","requiresSignup":true,"phoneVerify":false,"dataUsedForTraining":false,"speedNote":"","accessUrl":"https://openrouter.ai/models?max_price=0","sourceUrl":"https://openrouter.ai/docs/api-reference/limits","verifiedDate":"2026-06-28","note":"A single API gateway to dozens of free open-weight models. The current free pool (verified via the OpenRouter API) includes NVIDIA Nemotron Ultra/Super, OpenAI gpt-oss, Google Gemma 4, and Qwen3 Next; it is community-funded and can be throttled."},{"key":"opencode-zen","name":"OpenCode Zen","freeSummary":"Five free coding models (incl. MiMo V2.5, DeepSeek V4 Flash, Nemotron 3 Ultra trial, North Mini Code) via the OpenCode CLI and Desktop","requiresSignup":true,"phoneVerify":false,"dataUsedForTraining":true,"speedNote":"","accessUrl":"https://opencode.ai/docs/zen/","sourceUrl":"https://opencode.ai/docs/zen/","verifiedDate":"2026-06-28","note":"A curated coding-model gateway. On the free tier, data from MiMo V2.5, DeepSeek V4 Flash and North Mini Code may be used to improve the models; Nemotron 3 Ultra is trial-only via NVIDIA endpoints."},{"key":"nvidia-nim","name":"NVIDIA NIM (build.nvidia.com)","freeSummary":"Open models free at about 40 req/min after phone verification","requiresSignup":true,"phoneVerify":true,"dataUsedForTraining":false,"speedNote":"","accessUrl":"https://build.nvidia.com/","sourceUrl":"https://build.nvidia.com/","verifiedDate":"2026-06-28","note":"NVIDIA-hosted endpoints for open models including its own Nemotron line. Free for development after phone verification."},{"key":"groq","name":"Groq","freeSummary":"Open models free at roughly 1,000 req/day on larger models, up to 14,400 on small ones; 12K tokens/min","requiresSignup":true,"phoneVerify":false,"dataUsedForTraining":false,"speedNote":"Fastest-class inference on custom LPU hardware","accessUrl":"https://console.groq.com/","sourceUrl":"https://console.groq.com/docs/rate-limits","verifiedDate":"2026-06-28","note":"Built for speed. Free-tier request limits were tightened in 2026."},{"key":"cerebras","name":"Cerebras","freeSummary":"Open models free at 30 req/min, 14,400 req/day, 60K tokens/min","requiresSignup":true,"phoneVerify":false,"dataUsedForTraining":false,"speedNote":"Among the fastest output speeds available","accessUrl":"https://cloud.cerebras.ai/","sourceUrl":"https://inference-docs.cerebras.ai/support/pricing","verifiedDate":"2026-06-28","note":"Wafer-scale hardware delivers very high tokens-per-second. Free tier serves gpt-oss-120b and Llama models."},{"key":"cloudflare","name":"Cloudflare Workers AI","freeSummary":"10,000 neurons/day free across the model catalog (a usage credit, not a request cap)","requiresSignup":true,"phoneVerify":false,"dataUsedForTraining":false,"speedNote":"","accessUrl":"https://developers.cloudflare.com/workers-ai/","sourceUrl":"https://developers.cloudflare.com/workers-ai/platform/pricing/","verifiedDate":"2026-06-28","note":"Free daily neuron allocation runs open models (Llama, Qwen, gpt-oss) at the edge. Heavier models burn neurons faster."}],"unscored":[{"model":"North Mini Code","host":"OpenCode Zen / OpenRouter","reason":"A Cohere coding model offered free on OpenCode Zen and OpenRouter (cohere/north-mini-code:free), but Artificial Analysis does not score it, so it cannot be ranked on the same index as the rest."}]}