Alternatives to llama.cpp
For each task llama.cpp is listed in, the highest-ranked other options of the same kind, ranked only from public benchmarks.
Chat & assistants: runtimes & gateways
S
llama.cpp · Local · via Kimi K3
| Tier | Name | Approach | Runs | Rated by | Price | Skill |
|---|---|---|---|---|---|---|
| S |  Gemini Enterprise Agent Platform (formerly Vertex AI)Google Cloud | Runtimes & gateways | Cloud | via Gemini 4 Argon | — | advanced |
| S |  OpenRouterOpenRouter | Runtimes & gateways | Cloud | via Gemini 4 Argon | Free tier | intermediate |
| S |  RequestyRequesty | Runtimes & gateways | Cloud | via Gemini 4 Argon | — | intermediate |
| S |  Vercel AI GatewayVercel | Runtimes & gateways | Cloud | via Gemini 4 Argon | Free tier | intermediate |
| S |  Amazon BedrockAmazon Web Services | Runtimes & gateways | Cloud | via Claude Opus 5.5 | — | advanced |
| S |  DeepInfraDeepInfra | Runtimes & gateways | Cloud | via Claude Opus 5.5 | — | intermediate |
| S |  Microsoft FoundryMicrosoft | Runtimes & gateways | Cloud | via Claude Fable 5.1 | — | advanced |
| S |  OllamaOllama | Runtimes & gateways | Local | via Kimi K3 | Free | intermediate |
Writing: runtimes & gateways
A llama.cpp · Local · via Kimi K3
| Tier | Name | Approach | Runs | Rated by | Price | Skill |
|---|---|---|---|---|---|---|
| S | OpenRouter OpenRouter | Runtimes & gateways | Cloud | via Claude Opus 5.5 | Free tier | intermediate |
| A | Ollama Ollama | Runtimes & gateways | Local | via Kimi K3 | Free | intermediate |
Coding: runtimes & gateways
B llama.cpp · Local · via DeepSeek-V4.1-Flash-1017
| Tier | Name | Approach | Runs | Rated by | Price | Skill |
|---|---|---|---|---|---|---|
| S | Amazon Bedrock Amazon Web Services | Runtimes & gateways | Cloud | via Claude Opus 5.5 | — | advanced |
| S | DeepInfra DeepInfra | Runtimes & gateways | Cloud | via Claude Opus 5.5 | — | intermediate |
| S | Gemini Enterprise Agent Platform (formerly Vertex AI) Google Cloud | Runtimes & gateways | Cloud | via Claude Opus 5.5 | — | advanced |
| S | OpenRouter OpenRouter | Runtimes & gateways | Cloud | via Claude Opus 5.5 | Free tier | intermediate |
| S | Requesty Requesty | Runtimes & gateways | Cloud | via Claude Opus 5.5 | — | intermediate |
| S | Vercel AI Gateway Vercel | Runtimes & gateways | Cloud | via Claude Opus 5.5 | Free tier | intermediate |
| S | Microsoft Foundry Microsoft | Runtimes & gateways | Cloud | via GPT-6 Astra | — | advanced |
| B |  Novita AINovita AI | Runtimes & gateways | Cloud | via Qwen3.8-Max | — | intermediate |
Image understanding: runtimes & gateways
A llama.cpp · Local · via Gemma 4 (31B)
| Tier | Name | Approach | Runs | Rated by | Price | Skill |
|---|---|---|---|---|---|---|
| S | OpenRouter OpenRouter | Runtimes & gateways | Cloud | via Qwen3.8-Max | Free tier | intermediate |
| S | Ollama Ollama | Runtimes & gateways | Local | via GLM-5.3-Flash | Free | intermediate |
| A |  vLLMvLLM project | Runtimes & gateways | Local | via Gemma 4 (31B) | Free | advanced |
| A |  LocalAIEttore Di Giacinto (mudler) | Runtimes & gateways | Local | via Gemma 4 (31B) | Free | advanced |
| A |  SGLangSGLang project (LMSYS) | Runtimes & gateways | Local | via Gemma 4 (31B) | Free | advanced |
| A |  LMDeployInternLM (Shanghai AI Laboratory) | Runtimes & gateways | Local | via Gemma 4 (31B) | Free | advanced |
| A |  MLX-VLMBlaizzy (open-source project) | Runtimes & gateways | Local | via Gemma 4 (31B) | Free | advanced |
Tiers come only from public benchmarks; tools without a result of their own inherit the tier of the best model they offer. How we rank.