Loading…
Model Catalog
Every model usable through Hymn Router — free. Copy the model id into the model field in /api/v1/chat/completions. Context windows come from OpenRouter data; when there is no data it reads “unknown”.
7/8 active
The default model of each provider. Use hymn/auto for automatic fallback, a provider id (e.g. groq) to prioritize one provider, or the exact model id below.
Automatic router — tries all 9 providers in order until one succeeds. The default choice if you are not sure which model to use.
hymn/autoprovider: groq
Free: 30 RPM, 1,000 req/day, 200K tokens/day per model. No card.
openai/gpt-oss-120bprovider: nvidia
Free: ~40 RPM, 10K req/day per model. Trial/eval only, not for production.
nvidia/nemotron-3-ultra-550b-a55bprovider: gemini
Free: Flash-Lite 500 req/day; Flash 20 req/day. Free-tier prompts may be used by Google to improve its products.
gemini-3.5-flash-liteprovider: openrouter
Free: 20 RPM, 50 req/day (1,000/day after a one-time $10 top-up). :free models can change at any time.
nvidia/nemotron-3-ultra-550b-a55b:freeprovider: mistral
Free mode: $10 credit/month. No card.
mistral-small-latestprovider: kilo
No key: 200 req/hour per IP. The auto-router may send requests to providers that log prompts.
kilo-auto/freeprovider: llm7
Without a key you can call the turbo model. Free tokens at token.llm7.io: 60 RPM, 250 req/hour, 100K tokens/24h.
gpt-oss:20bprovider: pollinations · ovh
Keyless legacy lane, ~1 req/15 seconds. Verified working via node 2026-11-13.
gpt-oss-20bFree models from OpenRouter — send their id exactly as shown below in the model field to /v1 and the router forwards it to OpenRouter. Only models ending in :free are allowed. This list is live from OpenRouter (1-hour cache) and can change at any time.
Baidu: ERNIE-5.5-Flash-Preview (free)
context: 1,024,000 tokens
ERNIE 5.5 Flash Preview is the latest foundation model in the ERNIE family, with comprehensive improvements in coding, agentic capabilities, knowledge work, and…
baidu/ernie-5.5-flash-preview:freeCohere: North Mini Code (free)
context: 256,000 tokens
North Mini Code is Cohere's first agentic coding model and the debut of its North family. A sparse mixture-of-experts model with 30B total parameters and 3B act…
cohere/north-mini-code:freeDots Studio: Dots3-Note Preview (free)
context: 512,000 tokens
Dots3-Note Preview is an open-weight mixture-of-experts model from Dots Studio, with 16B active parameters out of 280B total. It is the lightest model in the Do…
dots-studio/dots-3-note-preview:freeGoogle: Gemma 4 26B A4B (free)
context: 262,144 tokens
Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token dur…
google/gemma-4-26b-a4b-it:freeGoogle: Gemma 4 31B (free)
context: 262,144 tokens
Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output. Features a 256K token context window, c…
google/gemma-4-31b-it:freeLiquidAI: LFM3-1.7B (free)
context: 262,144 tokens
A compact 1.7B-parameter text model with a 256K-token context, post-trained for agents, tool calling, and structured outputs. It reasons before it answers. Prom…
liquid/lfm-3-1.7b:freeNVIDIA: Nemotron 3 Nano Omni (free)
context: 256,000 tokens
NVIDIA Nemotron™ 3 Nano Omni is a 30B-A3B open multimodal model designed to function as a perception and context sub-agent in enterprise agent systems. It accep…
nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:freeNVIDIA: Nemotron 3 Ultra (free)
context: 1,000,000 tokens
NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid…
nvidia/nemotron-3-ultra-550b-a55b:freeNVIDIA: Nemotron 3.5 Lightning (free)
context: 1,000,000 tokens
NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, with 3B active parameters out of 30B total. It is suited for high-throughput agen…
nvidia/nemotron-3.5-lightning:freeNVIDIA: Nemotron 3.5 Super VL (free)
context: 1,000,000 tokens
NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total. It takes text, images, do…
nvidia/nemotron-3.5-super-vl:freeParticle AI: Particle-1 (free)
context: 1,048,576 tokens
Particle-1 is a token-efficient agentic coding model from Particle AI, post-trained on DeepSeek V4.1 Flash. It is suited for software engineering work such as t…
particle-ai/particle-1:freeRecursive: Flint (free)
context: 262,144 tokens
Flint is an open-weight mixture-of-experts model from Recursive, with about 2.5B active parameters out of 40B total, built by Recursive's autoresearch system Eu…
recursive/flint-v1.0:freeThinking Machines: Inkling 2 Small (free)
context: 1,048,576 tokens
Inkling 2 Small is an open-weight multimodal mixture-of-experts reasoning model from Thinking Machines Lab, with 17B active parameters out of 594B total. It is …
thinkingmachines/inkling-2-small:free