AgentToolbox

Free LLM models

Data updated · auto-synced daily from models.dev + OpenRouter

Every model below costs exactly $0 per token, aggregated from 7 providers. Generate ready-to-paste config for Claude Code, Codex CLI, or your own code.

"Free" means $0 input and $0 output — rate limits still apply per provider.

OpenRouter 17 free

Free (:free) models are rate-limited: 20 requests/min, plus 50 requests/day normally or 1000/day once you have bought ≥10 credits. The per-minute figure is a published tier default — the API does not report it per account.

ModelContextMax outCapabilitiesUpdatedConfig
Inkling (free)thinkingmachines/inkling:free1M256Ktools · reason · vision · open-weights2026-07
Inkling Small (free)thinkingmachines/inkling-small:free1M256Ktools · reason · vision · open-weights2026-07
MiniMax M3.1 Flash (free)minimax/minimax-m3.1-flash:free1M512Ktools · reason · vision2026-10
Nemotron 3 Ultra (free)nvidia/nemotron-3-ultra-550b-a55b:free1M64Ktools · reason · open-weights2026-06
Nemotron 3.5 Super VL (free)nvidia/nemotron-3.5-super-vl:free1M128Ktools · reason · vision · open-weights2026-10
Nemotron 3.5 Lightning (free)nvidia/nemotron-3.5-lightning:free1M64Ktools · reason · open-weights2026-08
Dots3-Note Preview (free)dots-studio/dots-3-note-preview:free500K450Ktools · reason · vision2026-08
Laguna XS 2.1 (free)poolside/laguna-xs-2.1:free256K32Ktools · reason · open-weights2026-07
Laguna S 2.1 (free)poolside/laguna-s-2.1:free256K32Ktools · reason · open-weights2026-07
Gemma 4 26B A4B (free)google/gemma-4-26b-a4b-it:free256K32Ktools · reason · vision · open-weights2026-04
Gemma 4 31B (free)google/gemma-4-31b-it:free256K32Ktools · reason · vision · open-weights2026-04
Xing4.0 29B A4B (free)xingchen-agi/xing4.0-29b-a4b:free256K230Ktools · reason · open-weights2026-10
Nemotron 3 Super (free)nvidia/nemotron-3-super-120b-a12b:free256K230Ktools · reason · open-weights2026-03
North Mini Code (free)cohere/north-mini-code:free250K63Ktools · reason · open-weights2026-06
Nemotron 3 Nano Omni (free)nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free250K64Ktools · reason · vision · open-weights2026-04
Nemotron 3.5 Content Safety (free)nvidia/nemotron-3.5-content-safety:free125K8Kreason · vision · open-weights2026-06
LFM2.5-2.6B (free)liquid/lfm-2.5-2.6b:free64K8Ktools · reason · open-weights2026-08

Nvidia 104 free

ModelContextMax outCapabilitiesUpdatedConfig
Kimi K3moonshotai/kimi-k31M128Ktools · reason · vision · open-weights2026-07
Step 5 Previewstepfun-ai/step-5-preview1M1Mtools · reason · vision · open-weights2026-09
Inklingthinkingmachines/inkling1M16Ktools · reason · vision · open-weights2026-07
Nemotron 3.5 Super VL 120B A12Bnvidia/nemotron-3.5-super-vl-120b-a12b1M8Ktools · reason · vision · open-weights2026-10
DeepSeek V4.1 Flashdeepseek-ai/deepseek-v4.1-flash1M375Ktools · reason · vision · open-weights2026-09
DeepSeek V4 Pro 0813deepseek-ai/deepseek-v4-pro-08131M375Ktools · reason · open-weights2026-08
DeepSeek V4 Flash 0731deepseek-ai/deepseek-v4-flash-07311M375Ktools · reason · open-weights2026-07
MiniMax-M3minimaxai/minimax-m31M16Ktools · reason · vision · open-weights2026-06
GLM-5.3-Flashz-ai/glm-5.3-flash1M128Ktools · reason · vision · open-weights2026-08
GLM-5.3z-ai/glm-5.31M128Ktools · reason · open-weights2026-08
GLM-5.2z-ai/glm-5.21M128Ktools · reason · open-weights2026-06
Kimi K2.6moonshotai/kimi-k2.6256K256Ktools · reason · vision · open-weights2026-04
Qwen3 Coder 480B A35B Instructqwen/qwen3-coder-480b-a35b-instruct256K65Ktools2025-07
Qwen3.5 122B-A10Bqwen/qwen3.5-122b-a10b256K64Ktools · reason · vision · open-weights2026-02
Qwen3.5-397B-A17Bqwen/qwen3.5-397b-a17b256K8Ktools · reason · vision · open-weights2026-02
Qwen3-Next-80B-A3B-Instructqwen/qwen3-next-80b-a3b-instruct256K16Ktools2025-09
Laguna XS 2.1poolside/laguna-xs-2.1256K16Ktools · reason · open-weights2026-07
Mistral Medium 3.5mistralai/mistral-medium-3.5-128b256K32Ktools · reason · vision · open-weights2026-04
Ministral 3 14B Instruct 2512mistralai/ministral-14b-instruct-2512256K16Ktools · vision · open-weights2025-12
Mistral Large 3 675B Instruct 2512mistralai/mistral-large-3-675b-instruct-2512256K256Ktools · vision · open-weights2025-12
Nemotron 3.5 Lightning 30B A3Bnvidia/nemotron-3.5-lightning-30b-a3b256K256Ktools · reason · open-weights2026-08
ByteDance-Seed/Seed-OSS-36B-Instructbytedance/seed-oss-36b-instruct256K256Ktools2025-11
Gemma-4-31B-ITgoogle/gemma-4-31b-it250K16Ktools · reason · vision · open-weights2026-04
Step 3.5 Flashstepfun-ai/step-3.5-flash250K16Ktools · reason · open-weights2026-02
Step 3.7 Flashstepfun-ai/step-3.7-flash250K16Ktools · reason · vision · open-weights2026-05
Nemotron 3 Nano Omninvidia/nemotron-3-nano-omni-30b-a3b-reasoning250K64Ktools · reason · vision · open-weights2026-04
DiffusionGemma 26B A4B ITgoogle/diffusiongemma-26b-a4b-it244K32Ktools · reason · vision · open-weights2026-07
MiniMax-M2.7minimaxai/minimax-m2.7200K128Ktools · reason · open-weights2026-04
Muse Glimmer 30Bmeta/muse-glimmer-30b128K128Ktools · reason · vision · open-weights2026-08
Gemma 3 12B ITgoogle/gemma-3-12b-it128K16Ktools · vision · open-weights2025-03
Gemma 3 4B ITgoogle/gemma-3-4b-it128K16Ktools · vision · open-weights2025-03
Phi-4-Minimicrosoft/phi-4-mini-instruct128K8Ktools · open-weights2025-09
GPT OSS 20Bopenai/gpt-oss-20b128K32Ktools · reason · open-weights2025-08
Mistral Medium 3mistralai/mistral-medium-3-instruct128K32Kvision2025-09
nvidia-nemotron-nano-9b-v2nvidia/nvidia-nemotron-nano-9b-v2128K128Ktools · reason · open-weights2025-08
Llama 3.1 Nemotron Nano 8B v1nvidia/llama-3.1-nemotron-nano-8b-v1128K16Ktools · reason · open-weights2025-03
Cosmos Reason2 8Bnvidia/cosmos-reason2-8b128K16Ktools · reason · vision · open-weights2025-12
Llama 3.3 Nemotron Super 49B v1.5nvidia/llama-3.3-nemotron-super-49b-v1.5128K64Ktools · reason · open-weights2025-07
Llama 3.3 Nemotron Super 49B v1nvidia/llama-3.3-nemotron-super-49b-v1128K64Ktools · reason · open-weights2025-03
nemotron-3-nano-30b-a3bnvidia/nemotron-3-nano-30b-a3b128K128Ktools · reason · open-weights2024-12
Llama 3.3 70b Instructmeta/llama-3.3-70b-instruct125K4Ktools · open-weights2024-11
Llama 3.1 70b Instructmeta/llama-3.1-70b-instruct125K4Ktools · open-weights2024-07
Llama Guard 4 12Bmeta/llama-guard-4-12b125K16Kvision · open-weights2026-04
Llama 3.2 1b Instructmeta/llama-3.2-1b-instruct125K4Ktools · open-weights2024-09
esmfoldmeta/esmfold125K8Kopen-weights2025-06
esm2-650mmeta/esm2-650m125K8Kopen-weights2025-03
Llama 4 Maverick 17b 128e Instructmeta/llama-4-maverick-17b-128e-instruct125K4Ktools · vision · open-weights2025-04
Llama 3.2 11b Vision Instructmeta/llama-3.2-11b-vision-instruct125K4Ktools · vision · open-weights2024-09
Llama-3.2-90B-Vision-Instructmeta/llama-3.2-90b-vision-instruct125K8Ktools · vision · open-weights2024-09
Qwen2.5 Coder 32b Instructqwen/qwen2.5-coder-32b-instruct125K4Ktools · open-weights2024-11
sarvam-msarvamai/sarvam-m125K8Ktools · open-weights2025-07
solar-10.7b-instructupstage/solar-10.7b-instruct125K8Ktools · open-weights2025-04
paligemmagoogle/google-paligemma125K8Kvision · open-weights2024-08
Gemma 3n E2b Itgoogle/gemma-3n-e2b-it125K4Ktools · vision · open-weights2025-06
Gemma 2 2b Itgoogle/gemma-2-2b-it125K4Ktools · open-weights2024-07
Gemma 3n E4b Itgoogle/gemma-3n-e4b-it125K4Ktools · vision · open-weights2025-06
Phi 4 Multimodalmicrosoft/phi-4-multimodal-instruct125K16K—2025-07
mistral-nemotronmistralai/mistral-nemotron125K8Ktools · open-weights2025-06
mistral-small-4-119b-2603mistralai/mistral-small-4-119b-2603125K8Ktools · reason · vision · open-weights2026-03
dracarys-llama-3.1-70b-instructabacusai/dracarys-llama-3.1-70b-instruct125K8Ktools · open-weights2025-05
nemotron-3-content-safetynvidia/nemotron-3-content-safety125K4Kopen-weights2026-04
gliner-piinvidia/gliner-pii125K4Kopen-weights2026-03
streampetrnvidia/streampetr125K8Kvision · open-weights2025-11
studiovoicenvidia/studiovoice125K8Kopen-weights2025-06
rerank-qa-mistral-4bnvidia/rerank-qa-mistral-4b125K4Kopen-weights2025-01
riva-translate-4b-instruct-v1_1nvidia/riva-translate-4b-instruct-v1.1125K4Kopen-weights2025-12
Llama 3.1 Nemotron Ultra 253Bnvidia/llama-3.1-nemotron-ultra-253b-v1125K16Ktools · reason · open-weights2025-04
sparsedrivenvidia/sparsedrive125K8Kvision · open-weights2025-07
nemotron-mini-4b-instructnvidia/nemotron-mini-4b-instruct125K8Ktools · open-weights2024-08
nemotron-content-safety-reasoning-4bnvidia/nemotron-content-safety-reasoning-4b125K4Kreason · open-weights2026-01
Nemotron 3.5 Content Safetynvidia/nemotron-3.5-content-safety125K8Kreason · vision · open-weights2026-06
llama-nemotron-rerank-vl-1b-v2nvidia/llama-nemotron-rerank-vl-1b-v2125K4Kvision · open-weights2026-03
usdcodenvidia/usdcode125K4K—2026-01
Nemotron Nano 12B v2 VLnvidia/nemotron-nano-12b-v2-vl125K125Ktools · reason · vision · open-weights2025-10
Llama 3.1 Nemotron 70B Instructnvidia/llama-3.1-nemotron-70b-instruct125K8Ktools · open-weights2025-04
llama-3.1-nemotron-safety-guard-8b-v3nvidia/llama-3.1-nemotron-safety-guard-8b-v3125K4Kopen-weights2025-10
nemotron-voicechatnvidia/nemotron-voicechat125K8Ktools · vision · open-weights2026-03
bevformernvidia/bevformer125K8Kvision · open-weights2025-07
Mistral-7B-Instruct-v0.3mistralai/mistral-7b-instruct-v0.364K64Ktools · open-weights2025-04
Mistral: Mixtral 8x22B Instructmistralai/mixtral-8x22b-instruct64K13Ktools · open-weights2024-04
FLUX.1-Kontext-devblack-forest-labs/flux_1-kontext-dev40K40Kvision · open-weights2025-08
FLUX.2 Klein 4Bblack-forest-labs/flux_2-klein-4b40K40Kopen-weights2026-01
Llama 3.2 3B Instructmeta/llama-3.2-3b-instruct32K31Kopen-weights2024-09
Mistral: Mixtral 8x7B Instructmistralai/mixtral-8x7b-instruct32K16Ktools · open-weights2026-03
Magistral Small 2506mistralai/magistral-small-250632K32K—2025-09
nv-embed-v1nvidia/nv-embed-v132K2Kopen-weights2025-07
Llama 3.1 Nemotron Nano VL 8B v1nvidia/llama-3.1-nemotron-nano-vl-8b-v132K16Ktools · reason · vision · open-weights2025-04
llama-nemotron-embed-vl-1b-v2nvidia/llama-nemotron-embed-vl-1b-v232K2Kvision · open-weights2026-02
llama-3_2-nemoretriever-300m-embed-v1nvidia/llama-3_2-nemoretriever-300m-embed-v132K2Kopen-weights2025-07
nv-embedcode-7b-v1nvidia/nv-embedcode-7b-v132K2Kopen-weights2025-05
Llama 3.1 8B Instructmeta/llama-3.1-8b-instruct16K4Ktools · open-weights2025-01
BGE M3baai/bge-m38K1Kopen-weights2026-04
FLUX.1-devblack-forest-labs/flux.1-dev4K0—2025-09
FLUX.1-schnellblack-forest-labs/flux_1-schnell770open-weights2026-02
Qwen Imageqwen/qwen-image00vision2025-08
Qwen Image Editqwen/qwen-image-edit00vision2025-08
Whisper Large v3openai/whisper-large-v304Kopen-weights2025-09
synthetic-video-detectornvidia/synthetic-video-detector04Kvision · open-weights2026-04
cosmos-transfer1-7bnvidia/cosmos-transfer1-7b04Kvision · open-weights2025-06
cosmos-transfer2.5-2bnvidia/cosmos-transfer2_5-2b04Kvision · open-weights2026-02
cosmos-predict1-5bnvidia/cosmos-predict1-5b04Kvision · open-weights2025-03
usdvalidatenvidia/usdvalidate04Kopen-weights2025-01
magpie-tts-zeroshotnvidia/magpie-tts-zeroshot04Kvision · open-weights2025-06
Active Speaker Detectionnvidia/active-speaker-detection04Kvision · open-weights2026-04

Groq 1 free

ModelContextMax outCapabilitiesUpdatedConfig
ALLaM-2-7ballam-2-7b4K4Kopen-weights2025-01

Mistral 2 free

ModelContextMax outCapabilitiesUpdatedConfig
Leanstral 1.5labs-leanstral-1-5-1256K125Ktools · reason · vision · open-weights2026-06
Devstral Small 2labs-devstral-small-2512250K250Ktools · vision · open-weights2025-12

Z.AI 3 free

ModelContextMax outCapabilitiesUpdatedConfig
GLM-4.7-Flashglm-4.7-flash195K128Ktools · reason · open-weights2026-01
GLM-4.5-Flashglm-4.5-flash128K96Ktools · reason2025-07
GLM-4.6V-Flashglm-4.6v-flash125K32Ktools · reason · vision · open-weights2025-12

Alibaba Token Plan 29 free

ModelContextMax outCapabilitiesUpdatedConfig
Qwen3.8 Maxqwen3.8-max1M128Ktools · reason · vision2026-08
DeepSeek V4.1 Flashdeepseek-v4.1-flash1M375Ktools · reason · vision · open-weights2026-09
Qwen3.7 Maxqwen3.7-max1M128Ktools · reason2026-05
DeepSeek V4 Pro 0813deepseek-v4-pro-08131M375Ktools · reason · open-weights2026-08
Qwen3.6 Flashqwen3.6-flash1M64Ktools · reason · vision2026-04
Qwen3.7 Plusqwen3.7-plus1M64Ktools · reason · vision2026-06
DeepSeek V4 Flash 0731deepseek-v4-flash-07311M375Ktools · reason · open-weights2026-07
DeepSeek V4 Flashdeepseek-v4-flash1M375Ktools · reason · open-weights2026-04
DeepSeek V4 Prodeepseek-v4-pro1M375Ktools · reason · open-weights2026-04
GLM-5.3glm-5.31M128Ktools · reason · open-weights2026-08
Qwen3.8 Flashqwen3.8-flash1M128Ktools · reason · vision2026-08
Qwen3.8 Max Previewqwen3.8-max-preview1M128Ktools · reason · vision2026-07
Qwen3.6 Plusqwen3.6-plus1M64Ktools · reason · vision2026-04
GLM-5.2glm-5.21M128Ktools · reason · open-weights2026-06
DeepSeek V4.1 Flash 1017deepseek-v4.1-flash-10171M375Ktools · reason · vision · open-weights2026-10
Kimi K2.5kimi-k2.5256K96Ktools · reason · vision · open-weights2026-01
Kimi K2.7 Codekimi-k2.7-code256K256Ktools · reason · vision · open-weights2026-06
Kimi K2.6kimi-k2.6256K256Ktools · reason · vision · open-weights2026-04
GLM-5.1glm-5.1198K125Ktools · reason · open-weights2026-04
GLM-5glm-5198K16Ktools · reason2026-02
MiniMax-M2.5MiniMax-M2.5192K32Ktools · reason · open-weights2026-02
DeepSeek V3.2deepseek-v3.2128K64Ktools · reason · open-weights2025-12
Qwen Image 2.0qwen-image-2.08K0—2026-03
Qwen Image 2.0 Proqwen-image-2.0-pro8K0—2026-03
Wan2.7 Image Prowan2.7-image-pro8K0—2026-05
Wan2.7 Imagewan2.7-image8K0—2026-05
HappyHorse 1.1 Text-to-Videohappyhorse-1.1-t2v00—2026-07
HappyHorse 1.1 Reference-to-Videohappyhorse-1.1-r2v00vision2026-07
HappyHorse 1.1 Image-to-Videohappyhorse-1.1-i2v00vision2026-07

OpenCode 9 free

Free models on the Zen endpoint — requires a (free) OpenCode API key; get one at opencode.ai.

ModelContextMax outCapabilitiesUpdatedConfig
Space Bunny Freespace-bunny-free1M512Ktools · reason · vision2026-09
Muse Spark 1.3 Freemuse-spark-1.3-contributor-free1M128Ktools · reason · vision2026-09
Fledge Alpha Freefledge-alpha-free1M128Ktools · reason · vision2026-10
Nemotron 3 Ultra Freenemotron-3-ultra-free1M125Ktools · reason · open-weights2026-06
LongCat 2.5 Preview Freelongcat-2.5-preview-free1M128Ktools · reason · vision2026-09
Nemotron 3.5 Lightning Freenemotron-3.5-lightning-free256K256Ktools · reason · open-weights2026-08
Ling 3.1 Flash Freeling-3.1-flash-free256K32Ktools · reason · open-weights2026-09
Ling 3.0 Flash Fin Freeling-3.0-flash-fin-free256K32Ktools · reason · open-weights2026-09
MiMo-V2.6-Flash Freemimo-v2.6-flash-free195K31Ktools · reason · vision · open-weights2026-09