Intelligence des modèles d'IA

Fournisseur · 2026-08-21

llmgateway-providers

10 modèles canoniques33 entrées au total (dérivés inclus)
ModèleEntrée / 1MSortie / 1MContexteFournisseursTags
InclusionAI Ling 3.0 Flash (DeepInfra)$0.060$0.180262K1tools · reasoning
GPT OSS 20B (Together AI)$0.050$0.200131K2tools · reasoning · open-weights
Cosmos 3 Super Reasoner (Nebius AI)$0.100$0.300262K1tools · reasoning · vision
Hermes 4 70B (Nebius AI)$0.130$0.400131K1tools · reasoning
Hy3 (DeepInfra)$0.140$0.580262K1tools · json · reasoning · open-weights
MiniCPM-V 4.5 (Nebius AI)$0.658$1.1132K1vision
Hermes 4 405B (Nebius AI)$1.00$3.00131K1tools · reasoning
o4 Mini (Azure)$1.10$4.40200K1tools · json · reasoning · vision
Fugu Ultra (Sakana AI)$5.00$30.001M1tools · reasoning · vision
o1 (Azure)$15.00$60.00200K1json · reasoning · vision
GLM-4 32B (0414-128k) (Z AI)dérivé$0.100$0.100128K1tools
Seed 1.6 Flash (250715) (ByteDance)dérivé$0.070$0.300256K1tools · reasoning · vision
Gemma 3 27B (Nebius AI)dérivé$0.100$0.300110K1vision
GLM-4.6V FlashX (Z AI)dérivé$0.040$0.400128K1tools · reasoning · vision
Qwen3 VL Flash (Alibaba Cloud)dérivé$0.050$0.400262K1tools · vision
GPT OSS 120B (ByteDance)dérivé$0.100$0.500128K7tools · reasoning · open-weights
DeepSeek V3.2 (ByteDance)dérivé$0.280$0.420131K1tools · reasoning · open-weights
MiniMax Text 01 (MiniMax)dérivé$0.200$1.101M1tools · reasoning
Qwen Coder Plus (Alibaba Cloud)dérivé$0.502$1.00131K1tools
Qwen Plus Latest (Alibaba Cloud)dérivé$0.400$1.201M1tools
DeepSeek V4 Flash (ByteDance)dérivé$0.440$1.321.05M2tools · reasoning · open-weights
Qwen3 235B A22B Instruct 2507 (Cerebras)dérivé$0.600$1.20262K1tools · open-weights
Llama 3.3 70B Instruct (Cerebras)dérivé$0.850$1.20128K1tools · open-weights
Seed 1.8 (251228) (ByteDance)dérivé$0.250$2.00256K1tools · reasoning · vision
Seed 1.6 (250915) (ByteDance)dérivé$0.250$2.00256K1tools · reasoning · vision
Seed 1.6 (250615) (ByteDance)dérivé$0.250$2.00256K1tools · reasoning · vision
Gemma 4 31B IT (Cerebras)dérivé$0.990$1.49131K1tools · reasoning · vision · open-weights
GLM-4.7 (ByteDance)dérivé$0.600$2.20200K2tools · reasoning · open-weights
GLM-5 (Alibaba Cloud)dérivé$0.573$2.58203K1tools · reasoning · open-weights
Kimi K2.5 (Alibaba Cloud)dérivé$0.574$3.01262K1tools · reasoning · vision · open-weights
Qwen3.5 397B A17B (Alibaba Cloud)dérivé$0.600$3.60262K1tools · reasoning · vision
DeepSeek V4 Pro (ByteDance)dérivé$1.32$3.961.05M2tools · reasoning · open-weights
GLM-5.2 (ByteDance)dérivé$1.40$4.401.02M2tools · reasoning · open-weights

Frequently asked questions

How many AI models does llmgateway-providers offer?

We track 10 canonical llmgateway-providers models plus 23 community fine-tunes / derivatives (excluded from the main table). The list is recomputed daily.

Which llmgateway-providers model is the cheapest?

InclusionAI Ling 3.0 Flash (DeepInfra) is currently the lowest-priced llmgateway-providers model, at $0.060 per 1M input tokens and $0.180 per 1M output tokens. For the full apples-to-apples list, see /pricing/cheapest-llm-api.

Which llmgateway-providers model has the largest context window?

Fugu Ultra (Sakana AI) leads at 1M tokens. This is the total of prompt + completion.

Which llmgateway-providers models support tool calling?

Multiple llmgateway-providers models support tool calling, with InclusionAI Ling 3.0 Flash (DeepInfra) being a popular pick. The capability column in the table above marks every model with llmgateway-providers tool-calling support.

Which llmgateway-providers models accept image input?

Cosmos 3 Super Reasoner (Nebius AI) accepts image input. Other vision-capable llmgateway-providers models are tagged 'vision' in the table above. See /capabilities/vision for a cross-vendor comparison.

What are the best alternatives to llmgateway-providers?

Depends on the use case. For raw cost savings, look at /pricing/cheapest-llm-api. For agent-oriented workloads, /best/best-ai-model-for-agents. For long-document workflows, /best/best-long-context-llm.

How fresh is this llmgateway-providers pricing data?

Daily. Our pipeline syncs every morning and rebuilds these pages on data change, so list-price moves and new model releases land within roughly 24 hours.

Dernière mise à jour :

Prices in USD per 1M tokens. Unknown means the provider does not publish per-token pricing.

Pricing and capabilities are refreshed daily and reconciled against each provider's official documentation. Always verify critical production decisions with the provider directly.