anytimelime — the free AI field report
Free AI Frontier: The Tool‑Capable Roster Shrinks to Six
Today’s probe reads like a quiet retreat. No new free models popped up to claim the sandbox; instead the only tool‑callers to survive the gauntlet are six survivors, and the SambaNova lineup that once offered a lone free passage is now uniformly behind a payment wall. The most striking fact is the narrowing of the free, tool‑ready camp: Nvidia’s Nemotron and Yolo‑Auto’s Qwen3.8‑27b are the only new entrants to the free‑tool club, while three models that were handy yesterday—inclusionai/ling-3.0-flash-vl:free, deepseek-ai/deepseek-v4-flash-0731 and liquid/lfm-2.5-2.6b:free—were knocked offline. In short, the free tier is contracting rather than expanding, and the pay‑walled SambaNova models (DeepSeek‑V3.1, DeepSeek‑V3.2, Meta‑Llama‑3.3‑70B‑Instruct, MiniMax‑M2.7, MiniMax‑M3, gemma‑4‑31B‑it) are today’s quiet background noise, still blocking the probe from even seeing if they could call tools.
What’s actually runnable right now
If you need a dependable, sub‑second workhorse for code‑style tasks, cohere/north-mini-code:free is still the fastest (0.6 s) and tool‑ready. For quick notes or concise summaries, dots-studio/dots-3-note-preview:free hangs in at 1.6 s—still respectable and fully tool‑capable. The three inclusionai flashes—ling‑3.0‑flash‑fin:free, ling‑3.0‑flash‑sante:free and ling‑3.0‑flash‑vl:free (the latter lost its tool flag today)—remain the sub‑second utility belt for finance, health and vision, respectively. Nvidia’s nvidia/nemotron-3-super-120b-a12b delivers a more heavyweight, 1.4 s response with tool support, while qwen3.8-27b from Yolo‑Auto is the slow‑but‑steady 2.3 s option that still accepts tools. The rest of the catalog is either empty, key‑less, or behind a payment prompt; none of them survive the probe’s tool‑call test.
What this means for builders on free tiers
The free, tool‑capable ecosystem is now a tight cluster of six models, and the churn is the story: a tool can appear overnight (as with Nemotron and Qwen3.8‑27b) and disappear just as fast. For anyone building pipelines that can’t rely on a single vendor, rotation is the only safe play. The SambaNova gatekeeping means you’ll need a payment method to tap those heavier models, but the OpenRouter‑wrapped six give you enough wiggle room to keep a prototype humming—provided you re‑probe before you lock in a dependency. In short, today’s free frontier is a shrinking playground where the only constant is change, and a well‑rotated, tool‑capable model list is the best defense against a silent lockout.
Clone github.com/fusionpoweredsoftware/anytimelime, set LIME_KEY to an OpenRouter key, and run ./lime.sh. It pulls today’s verified roster and rotates across the passing models, keeping your key on your own machine.
Today's passing roster
cohere/north-mini-code:free— openrouter · 0.6s · agentic 3/5inclusionai/ling-3.0-flash-fin:free— openrouter · 0.8s · agentic 3/5inclusionai/ling-3.0-flash-sante:free— openrouter · 0.9s · agentic 3/5nvidia/nemotron-3-super-120b-a12b— NVIDIA NIM · 1.4s · agentic 3/5dots-studio/dots-3-note-preview:free— openrouter · 1.6s · agentic 3/5qwen3.8-27b— Yolo-Auto · 2.3s · agentic 3/5
The measured table
| latency | model | vendor | tool call | free now | agentic | yesterday | 3 days ago | a week ago |
|---|---|---|---|---|---|---|---|---|
| 0.7 | gemini-2.5-flash | Google Gemini (OpenAI-compat) | untested | DOWN | — | · | · | · |
| — | meta-llama/Llama-3.3-70B-Instruct | Nebius (Token Factory) | untested | DOWN | — | · | · | · |
| 0.6 | deepseek-v4-flash:free | UnoRouter | untested | DOWN | — | · | · | · |
| 0.2 | z-ai/glm-5.2:free | openrouter | untested | DOWN | — | · | · | · |
| 0.4 | meganova-ai/manta-flash-1.0 | MegaNova AI | untested | DOWN | — | · | ✗ | · |
| 0.3 | mistral-small-latest | Mistral AI | untested | DOWN | — | · | · | · |
| 1.4 | nvidia/nemotron-3-super-120b-a12b | NVIDIA NIM | yes | FREE | 3/5 | · | · | ✓ |
| 0.3 | aisingapore/Qwen-SEA-LION-v4.5-27B-IT | SEA-LION (AI Singapore) | yes (past) | DOWN | — | · | · | ✓ |
| 2.3 | qwen3.8-27b | Yolo-Auto | yes | FREE | 3/5 | · | · | · |
| 0.2 | 01-ai/yi-large | nvidia | untested | DOWN | — | ✗ | ✗ | ✗ |
| 0.6 | cohere/north-mini-code:free | openrouter | yes | FREE | 3/5 | ✓ | ✓ | ✓ |
| 0.3 | DeepSeek-V3.1 | sambanova | untested | CARD | — | ✗ | ✗ | ✗ |
| 0.1 | adept/fuyu-8b | nvidia | untested | DOWN | — | ✗ | ✗ | ✗ |
| 1.6 | dots-studio/dots-3-note-preview:free | openrouter | yes | FREE | 3/5 | ✓ | ✓ | ✓ |
| 0.3 | DeepSeek-V3.2 | sambanova | untested | CARD | — | ✗ | ✗ | ✗ |
| 0.2 | ai21labs/jamba-1.5-large-instruct | nvidia | untested | DOWN | — | ✗ | ✗ | ✗ |
| 0.2 | google/gemma-4-26b-a4b-it:free | openrouter | yes (past) | DOWN | — | ✗ | ✗ | ✗ |
| 0.3 | Meta-Llama-3.3-70B-Instruct | sambanova | untested | CARD | — | ✗ | ✗ | ✗ |
| 0.2 | aisingapore/sea-lion-7b-instruct | nvidia | untested | DOWN | — | ✗ | ✗ | ✗ |
| 0.3 | google/gemma-4-31b-it:free | openrouter | yes (past) | DOWN | — | ✗ | ✗ | ✗ |
| 0.3 | MiniMax-M2.7 | sambanova | untested | CARD | — | ✗ | ✗ | ✗ |
| 0.1 | bigcode/starcoder2-15b | nvidia | untested | DOWN | — | ✗ | ✗ | ✗ |
| 0.8 | inclusionai/ling-3.0-flash-fin:free | openrouter | yes | FREE | 3/5 | ✓ | ✓ | ✓ |
| 0.3 | MiniMax-M3 | sambanova | untested | CARD | — | ✗ | ✗ | ✗ |
| 0.1 | databricks/dbrx-instruct | nvidia | untested | DOWN | — | ✗ | ✗ | ✗ |
| 0.9 | inclusionai/ling-3.0-flash-sante:free | openrouter | yes | FREE | 3/5 | ✓ | ✓ | ✓ |
| 0.3 | gemma-4-31B-it | sambanova | untested | CARD | — | ✗ | ✗ | ✗ |
| 0.1 | deepseek-ai/deepseek-coder-6.7b-instruct | nvidia | untested | DOWN | — | ✗ | ✗ | ✗ |