anytimelime — the free AI field report

2026-10-05 · generated by a free model interviewing free models · anytimelime, formerly Project42 · get the engine →

What's free in AI right now — 2026-10-05

This morning the free tier has a ghost in the machine: the once‑tool‑capable Qwen/Qwen3-Coder-480B-A35B-Instruct vanished behind a 60‑second timeout, while a quiet paywall swallowed inclusionai/ling-3.1-flash (OpenRouter) and kilo (Kilo Gateway). In a week that promised stability, those two shifts are the standout drama.

Running the free pool today feels like navigating a timed‑run of dependable workhorses and fickle side‑doors. If you need a tool‑enabled model under a second, the Pollinations openai endpoint (0.3 s) and the OpenRouter openai/gpt-oss-120b:free (1.0 s) stay fast and functional. The newly tool‑capable MiniMax-M2.7-Official (2.5 s) and its sibling MiniMaxAI/MiniMax-M2.7 (2.1 s) deliver modest latency with a rare tool‑use flag—useful for quick reasoning bursts. For anyone chasing near‑real‑time responses without burning credits, the NVIDIA nemotron-3-super-120b-a12b (0.6 s) and the Hugging Face meta-llama/Llama-3.3-70B-Instruct (1.6 s) remain steady. A few high‑latency free tools survive the shuffle, like the 5‑second nvidia/nemotron-3.5-lightning:free and the 6‑second liquid/lfm-2.5-2.6b:free

For builders who rely on free tiers, today’s churn is a reminder that the roster is a living circuit, not a static shelf. One model that answered tools yesterday can be silent tomorrow, and a “free” label can be stripped without warning. The spread from a few hundred milliseconds to minute‑long waits means any production script should prioritize speed first, then keep a fallback queue in motion before the next lock‑out hits. Rotate early, rotate often, and treat the free pool as a rotating relay rather than a permanent grid.

Tech Talk Tuesday — 2026-10-04 to 2026-10-05

This week no new models claimed tool capability—gains: none

The net effect is a tighter, more selective free pool—good for performance‑sensitive work, bad for those who counted on a broader set of tools. The trend continues: paywalls and rate limits are the new normal, and only the most resilient models stay free‑and‑tool‑ready.

To spin your own free‑tier spinner, grab the repo from github.com/fusionpoweredsoftware/anytimelime, drop your OpenRouter key into LIME_KEY, and run ./lime.sh. It pulls today’s verified roster and rotates across the passing models; your key never leaves your terminal.

Today's passing roster

The measured table

Click any model for its endpoint and ready-to-run snippets in curl, Python, Go, Rust, and more.

latencymodelvendortool callfree nowagenticyesterday3 days agoa week ago
—ai21AI21 LabsuntestedDOWN—···
—alibaba-dashscopeAlibaba Cloud Model Studio (DashScope)untestedDOWN—···
0.4chutesChutes AIuntestedDOWN—···
0.2cloudflare-workers-aiCloudflare Workers AIuntestedDOWN—···
—deepinfraDeepInfrauntestedDOWN—···
—fireworksFireworks AIuntestedDOWN—···
0.2google-aistudioGoogle AI Studio (Gemini)untestedDOWN—···
0.2huggingface-routerHugging Face Inference RouteruntestedDOWN—···
—internlmInternLM (Intern AI)untestedDOWN—···
0.5kiloKilo GatewayuntestedCARD—···
0.3mistralMistral AI (La Plateforme)untestedDOWN—···
—modelscopeModelScope API-InferenceuntestedDOWN—···
0.5novitaNovita AIuntestedDOWN—···
0.1ollama-cloudOllama ClouduntestedDOWN—···
0.9openclawOpenClaw AIuntestedDOWN—···
0.6sambanovaSambaNova ClouduntestedDOWN—···
—scalewayScaleway Generative APIsuntestedDOWN—···
0.4sea-lionSEA-LIONuntestedDOWN—···
0.5unorouterUnoRouteruntestedDOWN—···
—zhipuZhipu AI (GLM / z.ai)untestedDOWN—···
0.2aion-labsAION LabsuntestedDOWN—···
0.2nvidia-nimNVIDIA NIMuntestedDOWN—···
0.7siliconflowSiliconFlowuntestedDOWN—···
0.2nvidia/vilanvidiauntestedDOWN—···
0.101-ai/yi-largenvidiauntestedDOWN—✗✗✗
0.3DeepSeek-V3.1sambanovauntestedDOWN—✗✗✗
2.5MiniMax-M2.7-OfficialPooled AIyesFREE3/5✓✓✓
2.1MiniMaxAI/MiniMax-M2.7Gonka BrokeryesFREE3/5✓✓✓
60.1Qwen/Qwen3-Coder-480B-A35B-InstructHuggingFace Routeryes (past)DOWN—✓✓✓
4.6agnes-3.0-flashAgnes AIyesFREE3/5✓✓✓
0.4aion-labs/aion-3.0AION Labsyes (past)DOWN—✓✗✓
0.2aisingapore/Qwen-SEA-LION-v4.5-27B-ITSEA-LION (AI Singapore)yes (past)DOWN—✗✗✗
2.0apodex/apodex-1.1-mini:freeopenrouteryesFREE3/5✓✓·
0.3codestral-latestMistralyes (past)DOWN—✗✗✗
0.7deepseek-ai/DeepSeek-V3-0324HuggingFace Inference RouteryesFREE3/5✓✓✓
0.3deepseek-ai/DeepSeek-V3.1SiliconFlowyes (past)DOWN—✗✗✗
0.1deepseek-ai/deepseek-v4-flash-0731NVIDIA NIMyes (past)DOWN—✗✗✗
0.5glm-4.5-flashZhipu AI (Z.ai)yes (past)DOWN—✗✗✗
33.4glm-4.7-flashZhipu AI (Z.ai)yesFREE3/5✓✗✓
1.5intern-latestIntern AIyes (past)DOWN—✗✗✗
1.6meta-llama/Llama-3.3-70B-InstructHugging Face InferenceyesFREE3/5✓✓✓
0.4minimax/minimax-m3:freexKiro AIyes (past)DOWN—✗✓✗
1.9nemotron-3-super-120b-a12b:freeNaga AIyesFREE3/5✓✓✓
3.6north-mini-codeTokenReplyyesFREE3/5✓✓✓
1.3openai-fastPollinationsyesFREE3/5✓✓✓
1.0openai/gpt-oss-120b:freeFastRouteryesFREE3/5✓✓✓
0.3openaiPollinationsyesFREE3/5✓✓✓
1.7qwen/qwen3.5-4b-freeMixlayeryesFREE3/5✓✓✓
1.4qwen3-8bFree.aiyesFREE3/5✓✓✓
0.2qwen3.8-27bYolo-Autoyes (past)DOWN—✗✗✗
60.1tokeness/freeTokenessyes (past)DOWN—✗✓✗
1.1typhoon-v2.5-30b-a3b-instructTyphoon (SCB 10X)yesFREE3/5✓✓✓
0.1adept/fuyu-8bnvidiauntestedDOWN—✗✗✗
0.3DeepSeek-V3.2sambanovauntestedDOWN—✗✗✗
0.6deepseek-ai/DeepSeek-V4-Flash-0731Gonka Brokeryes (past)DOWN—✗✗✗
1.5cohere/north-mini-code:freeopenrouteryesFREE3/5✓✓✓
0.4inclusionAI/Ling-flash-2.0SiliconFlowyes (past)DOWN—✗✗✗
33.4glm-4.7-flashZhipu AI (Z.ai)yesFREE3/5✓✗✓
2.7intern-s2-previewIntern AIyesFREE3/5✓✓✓
0.3openaiPollinationsyesFREE3/5✓✓✓
1.0openai/gpt-oss-20b:freeFastRouteryesFREE3/5✓✓✓
0.2ai21labs/jamba-1.5-large-instructnvidiauntestedDOWN—✗✗✗
0.4Meta-Llama-3.3-70B-InstructsambanovauntestedDOWN—✗✗✗
2.1zai-org/GLM-5.3-FlashGonka BrokeryesFREE3/5✓✓✓
2.1dots-studio/dots-3-note-preview:freeopenrouteryesFREE3/5✓✓✓
0.2aisingapore/sea-lion-7b-instructnvidiauntestedDOWN—✗✗✗
0.3MiniMax-M3sambanovauntestedDOWN—✗✗✗
0.3google/gemma-4-26b-a4b-it:freeopenrouteryes (past)DOWN—✗✗✗
0.1bigcode/starcoder2-15bnvidiauntestedDOWN—✗✗✗
0.3gemma-4-31B-itsambanovauntestedDOWN—✗✗✗
0.3google/gemma-4-31b-it:freeopenrouteryes (past)DOWN—✗✗✗
0.2databricks/dbrx-instructnvidiauntestedDOWN—✗✗✗
0.3gpt-oss-120bsambanovauntestedDOWN—✗✗✗
1.2inclusionai/ling-3.0-flash-sante:freeopenrouteryesFREE3/5✓✓✓
0.2deepseek-ai/deepseek-coder-6.7b-instructnvidiauntestedDOWN—✗✗✗
0.2inclusionai/ling-3.1-flashopenrouteruntestedCARD—✗··
60.1deepseek-ai/deepseek-v4.1-flashnvidiauntestedDOWN—✗✗✗
6.1liquid/lfm-2.5-2.6b:freeopenrouteryesFREE3/5✓✓✓
0.2google/codegemma-1.1-7bnvidiauntestedDOWN—✗✗✗
0.3nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:freeopenrouteryes (past)DOWN—✓✗✓
0.2google/codegemma-7bnvidiauntestedDOWN—✗✗✗
0.8nvidia/nemotron-3-super-120b-a12b:freeopenrouteryesFREE3/5✓✓✗
0.2google/deplotnvidiauntestedDOWN—✗✗✗
0.4nvidia/nemotron-3-ultra-550b-a55b:freeopenrouteryes (past)DOWN—✓✓✓
0.3google/diffusiongemma-26b-a4b-itnvidiayesFREE3/5✓✓✓
0.1nvidia/nemotron-3.5-content-safety:freeopenrouteruntestedDOWN—✗✗✗
0.2google/gemma-2bnvidiauntestedDOWN—✗✗✗
4.1nvidia/nemotron-3.5-lightning:freeopenrouteryesFREE3/5✓✓✓
0.1google/gemma-3-12b-itnvidiauntestedDOWN—✗✗✗
1.4openrouter/freeopenrouteryesFREE3/5✓✓✓
0.1google/gemma-3-4b-itnvidiauntestedDOWN—✗✗✗
0.4poolside/laguna-s-2.1:freeopenrouteryes (past)DOWN—✓✗✗
58.2google/gemma-4-31b-itnvidiayesFREE3/5✓✓✓
0.3poolside/laguna-xs-2.1:freeopenrouteryes (past)DOWN—✓✓✗
0.1google/recurrentgemma-2bnvidiauntestedDOWN—✗✗✗
0.1ibm/granite-3.0-3b-a800m-instructnvidiauntestedDOWN—✗✗✗
0.2ibm/granite-3.0-8b-instructnvidiauntestedDOWN—✗✗✗
0.2ibm/granite-34b-code-instructnvidiauntestedDOWN—✗✗✗
0.1ibm/granite-8b-code-instructnvidiauntestedDOWN—✗✗✗
0.2meta/codellama-70bnvidiauntestedDOWN—✗✗✗
0.7meta/llama-3.2-11b-vision-instructnvidiayesFREE3/5✓✓✓
48.0meta/llama-3.2-90b-vision-instructnvidiayesFREE3/5✓✓✓
0.2meta/llama2-70bnvidiauntestedDOWN—✗✗✗
7.0meta/muse-glimmer-30bnvidiayesFREE3/5✓✓✓
0.2microsoft/kosmos-2nvidiauntestedDOWN—✗✗✗
0.1microsoft/phi-3-vision-128k-instructnvidiauntestedDOWN—✗✗✗
0.2microsoft/phi-3.5-moe-instructnvidiauntestedDOWN—✗✗✗
0.2mistralai/codestral-22b-instruct-v0.1nvidiauntestedDOWN—✗✗✗
0.2mistralai/mistral-7b-instruct-v0.3nvidiauntestedDOWN—✗✗✗
0.2mistralai/mistral-large-2-instructnvidiauntestedDOWN—✗✗✗
0.2mistralai/mistral-largenvidiauntestedDOWN—✗✗✗
0.2mistralai/mixtral-8x22b-v0.1nvidiauntestedDOWN—✗✗✗
0.2moonshotai/kimi-k2.6nvidiauntestedDOWN—✗✗✗
18.1moonshotai/kimi-k3nvidiayesFREE3/5✓✓✗
0.1nv-mistralai/mistral-nemo-12b-instructnvidiauntestedDOWN—✗✗✗
0.2nvidia/ai-synthetic-video-detectornvidiauntestedDOWN—✗✗✗
0.1nvidia/cosmos-reason2-8bnvidiauntestedDOWN—✗✗✗
4.9nvidia/ising-calibration-1.5-31bnvidiatext-onlyFREE—✓✓✓
0.1nvidia/llama-3.1-nemotron-51b-instructnvidiauntestedDOWN—✗✗✗
0.2nvidia/llama-3.1-nemotron-70b-instructnvidiauntestedDOWN—✗✗✗
0.1nvidia/llama-3.1-nemotron-ultra-253b-v1nvidiauntestedDOWN—✗✗✗
0.2nvidia/llama3-chatqa-1.5-70bnvidiauntestedDOWN—✗✗✗
0.2nvidia/mistral-nemo-minitron-8b-8k-instructnvidiauntestedDOWN—✗✗✗
0.2nvidia/nemotron-3-nano-omni-30b-a3b-reasoningnvidiayes (past)DOWN—✗✗✓
0.6nvidia/nemotron-3-super-120b-a12bnvidiayesFREE3/5✓✓✓
2.2nvidia/nemotron-3-ultra-550b-a55bnvidiayesFREE3/5✓✓✓
0.2nvidia/nemotron-3.5-content-safetynvidiauntestedDOWN—✗✗✗
5.1nvidia/nemotron-3.5-lightning-30b-a3bnvidiayesFREE3/5✓✓✓
0.2nvidia/nemotron-4-340b-instructnvidiauntestedDOWN—✗✗✗
0.1nvidia/nemotron-4-340b-rewardnvidiauntestedDOWN—✗✗✗
0.1nvidia/nemotron-nano-3-30b-a3bnvidiauntestedDOWN—✗✗✗
0.2nvidia/nemotron-parse-2.0nvidiauntestedDOWN—✗✗✗
0.2nvidia/nemotron-parsenvidiauntestedDOWN—✗✗✗
0.1nvidia/neva-22bnvidiauntestedDOWN—✗✗✗
0.1nvidia/nvclipnvidiauntestedDOWN—✗✗✗
11.9nvidia/riva-translate-4b-instruct-v2nvidiauntestedDOWN—✗✗✗
0.1nvidia/riva-translate-4b-instructnvidiauntestedDOWN—✗✗✗

35 made a tool call, 1 answered in text only, of 137 discovered. A fail only means the endpoint wasn't free to use right then — it demanded a payment method, or it was down or erroring — so tool-call ability stayed untested, not disproven; a model that passed in a past report keeps its yes. Numbers are single-run measurements from one location — your mileage will vary. Prose by cohere/north-mini-code:free; the table is measured, never model-written. Reproduce with ./free-scan.sh from the engine repo.