{"apiVersion":"v1","directorySlug":"poorpaul-mcp-c6ce15878d","title":"Open LLM Inference Benchmarks","description":"Explore open benchmarks for local AI inference on consumer hardware. Curated leaderboards, interactive charts, and analysis from Poor Paul's Benchmark.","pageUrl":"https://www.mcpbundles.com/skills/poorpaul-mcp-c6ce15878d","connectPath":"importable","connectAction":"import_server","catalogListing":false,"bundleSlug":null,"listingKind":"remote_mcp","kind":"provider","transportUrl":"https://mcp.poorpaul.dev/mcp","hostedMcpUrl":null,"toolCount":14,"toolInventoryStatus":"observed","refreshedAt":"2026-11-03T13:26:00.516172+00:00","registryName":"dev.poorpaul/poorpaul-mcp-c6ce15878d","registryVersion":"2026.11.3","registryEligible":true,"tools":[{"name":"list_tested_configs","description":"List every tested GPU, model, quantization, and runner type in the PPB dataset.","inputSchema":{"type":"object","properties":{},"additionalProperties":false},"logicalName":"list_tested_configs","displayName":"list_tested_configs","title":"list_tested_configs","annotations":{"untrustedContentHint":true,"readOnlyHint":true}},{"name":"query_ppb_results","description":"Filter raw benchmark rows from PPB.","inputSchema":{"type":"object","properties":{"limit":{"type":"integer","default":50},"model":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"backend":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"gpu_name":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"run_after":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"run_before":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"exact_model":{"type":"boolean","default":false},"runner_type":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"vram_gb_max":{"anyOf":[{"type":"number"},{"type":"null"}],"default":null},"vram_gb_min":{"anyOf":[{"type":"number"},{"type":"null"}],"default":null},"quantization":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"unified_memory":{"anyOf":[{"type":"boolean"},{"type":"null"}],"default":null},"llm_flags_label":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"concurrent_users":{"anyOf":[{"type":"integer"},{"type":"null"}],"default":null}},"additionalProperties":false},"logicalName":"query_ppb_results","displayName":"query_ppb_results","title":"query_ppb_results","annotations":{"untrustedContentHint":true,"readOnlyHint":true}},{"name":"recommend_quantization","description":"Recommend the best quantization for a given GPU VRAM budget and user count.","inputSchema":{"type":"object","required":["concurrent_users"],"properties":{"model":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"gpu_name":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"priority":{"enum":["quality","speed","balance","efficiency"],"type":"string","default":"balance"},"gpu_vram_gb":{"anyOf":[{"type":"number"},{"type":"null"}],"default":null},"concurrent_users":{"type":"integer"}},"additionalProperties":false},"logicalName":"recommend_quantization","displayName":"recommend_quantization","title":"recommend_quantization","annotations":{"untrustedContentHint":true,"readOnlyHint":true}},{"name":"get_gpu_headroom","description":"Sanity-check VRAM usage for a specific (gpu, quant, model, users) config.","inputSchema":{"type":"object","required":["gpu_name","quantization","model"],"properties":{"model":{"type":"string"},"gpu_name":{"type":"string"},"quantization":{"type":"string"},"concurrent_users":{"type":"integer","default":1}},"additionalProperties":false},"logicalName":"get_gpu_headroom","displayName":"get_gpu_headroom","title":"get_gpu_headroom","annotations":{"untrustedContentHint":true,"readOnlyHint":true}},{"name":"compare_quants_quantitative","description":"Compare quantitative benchmark scores across quantizations for a model.","inputSchema":{"type":"object","required":["model"],"properties":{"model":{"type":"string"},"gpu_name":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"runner_type":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"llm_flags_label":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"concurrent_users":{"anyOf":[{"type":"integer"},{"type":"null"}],"default":null}},"additionalProperties":false},"logicalName":"compare_quants_quantitative","displayName":"compare_quants_quantitative","title":"compare_quants_quantitative","annotations":{"untrustedContentHint":true,"readOnlyHint":true}},{"name":"get_combined_scores","description":"Get both quantitative (speed/VRAM) and qualitative (accuracy/quality) scores\nfor a single (gpu, model, quant) configuration in one call.","inputSchema":{"type":"object","required":["model","quantization"],"properties":{"model":{"type":"string"},"gpu_name":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"quantization":{"type":"string"}},"additionalProperties":false},"logicalName":"get_combined_scores","displayName":"get_combined_scores","title":"get_combined_scores","annotations":{"untrustedContentHint":true,"readOnlyHint":true}},{"name":"rank_by_priority","description":"Rank all tested quantizations for a model by a composite score.","inputSchema":{"type":"object","required":["model"],"properties":{"limit":{"type":"integer","default":20},"model":{"type":"string"},"gpu_name":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"priority":{"enum":["speed","quality","balance","efficiency"],"type":"string","default":"balance"}},"additionalProperties":false},"logicalName":"rank_by_priority","displayName":"rank_by_priority","title":"rank_by_priority","annotations":{"untrustedContentHint":true,"readOnlyHint":true}},{"name":"recommend_hardware","description":"Recommend the best GPU for running a specific model.","inputSchema":{"type":"object","required":["target_model"],"properties":{"priority":{"enum":["speed","efficiency","value"],"type":"string","default":"value"},"budget_usd":{"anyOf":[{"type":"number"},{"type":"null"}],"default":null},"target_model":{"type":"string"},"concurrent_users":{"type":"integer","default":1},"target_quantization":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null}},"additionalProperties":false},"logicalName":"recommend_hardware","displayName":"recommend_hardware","title":"recommend_hardware","annotations":{"untrustedContentHint":true,"readOnlyHint":true}},{"name":"explain_result","description":"Explain why a benchmark result is what it is.","inputSchema":{"type":"object","required":["gpu_name","model","quantization"],"properties":{"model":{"type":"string"},"n_ctx":{"anyOf":[{"type":"integer"},{"type":"null"}],"default":null},"gpu_name":{"type":"string"},"quantization":{"type":"string"},"concurrent_users":{"type":"integer","default":1}},"additionalProperties":false},"logicalName":"explain_result","displayName":"explain_result","title":"explain_result","annotations":{"untrustedContentHint":true,"readOnlyHint":true}},{"name":"get_qualitative_summary","description":"Get all available qualitative benchmark scores for a model, optionally filtered\nto a specific quantization and/or GPU.","inputSchema":{"type":"object","required":["model"],"properties":{"model":{"type":"string"},"gpu_name":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"quantization":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null}},"additionalProperties":false},"logicalName":"get_qualitative_summary","displayName":"get_qualitative_summary","title":"get_qualitative_summary","annotations":{"untrustedContentHint":true,"readOnlyHint":true}},{"name":"query_qualitative_results","description":"Query qualitative benchmark results with optional filters.","inputSchema":{"type":"object","properties":{"limit":{"type":"integer","default":50},"model":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"gpu_name":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"runner_type":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"quantization":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"min_mt_bench_score":{"anyOf":[{"type":"number"},{"type":"null"}],"default":null},"min_context_rot_score":{"anyOf":[{"type":"number"},{"type":"null"}],"default":null},"min_overall_tool_accuracy":{"anyOf":[{"type":"number"},{"type":"null"}],"default":null}},"additionalProperties":false},"logicalName":"query_qualitative_results","displayName":"query_qualitative_results","title":"query_qualitative_results","annotations":{"untrustedContentHint":true,"readOnlyHint":true}},{"name":"get_context_rot_breakdown","description":"Get detailed context-rot (long-context recall) scores broken down by\ncontext length, insertion depth, and needle type.","inputSchema":{"type":"object","required":["model","quantization"],"properties":{"model":{"type":"string"},"gpu_name":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"quantization":{"type":"string"}},"additionalProperties":false},"logicalName":"get_context_rot_breakdown","displayName":"get_context_rot_breakdown","title":"get_context_rot_breakdown","annotations":{"untrustedContentHint":true,"readOnlyHint":true}},{"name":"get_tool_accuracy_breakdown","description":"Get detailed tool-call accuracy metrics for a model+quant.","inputSchema":{"type":"object","required":["model","quantization"],"properties":{"model":{"type":"string"},"gpu_name":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"quantization":{"type":"string"}},"additionalProperties":false},"logicalName":"get_tool_accuracy_breakdown","displayName":"get_tool_accuracy_breakdown","title":"get_tool_accuracy_breakdown","annotations":{"untrustedContentHint":true,"readOnlyHint":true}},{"name":"compare_quants_qualitative","description":"Compare qualitative benchmark scores across quantizations for a model.","inputSchema":{"type":"object","required":["model"],"properties":{"model":{"type":"string"},"gpu_name":{"anyOf":[{"type":"string"},{"type":"null"}],"default":null},"quantizations":{"anyOf":[{"type":"array","items":{"type":"string"}},{"type":"null"}],"default":null}},"additionalProperties":false},"logicalName":"compare_quants_qualitative","displayName":"compare_quants_qualitative","title":"compare_quants_qualitative","annotations":{"untrustedContentHint":true,"readOnlyHint":true}}],"toolSource":"directory_preview","agentAction":"both","installLinks":[{"id":"cursor","label":"Cursor","href":"cursor://anysphere.cursor-deeplink/mcp/install?name=Open-LLM-Inference-Benchmarks&config=eyJ1cmwiOiAiaHR0cHM6Ly9tY3AucG9vcnBhdWwuZGV2L21jcCJ9"},{"id":"vscode","label":"VS Code","href":"vscode:mcp/install?%7B%22name%22%3A%22Open-LLM-Inference-Benchmarks%22%2C%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fmcp.poorpaul.dev%2Fmcp%22%7D"},{"id":"vscode-insiders","label":"VS Code Insiders","href":"vscode-insiders:mcp/install?%7B%22name%22%3A%22Open-LLM-Inference-Benchmarks%22%2C%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fmcp.poorpaul.dev%2Fmcp%22%7D"},{"id":"claude-web","label":"Claude.ai","href":"https://claude.ai/customize/connectors?modal=add-custom-connector&connectorName=Open+LLM+Inference+Benchmarks&connectorUrl=https%3A%2F%2Fmcp.poorpaul.dev%2Fmcp"},{"id":"codex","label":"Codex","copyText":"codex mcp add \"Open-LLM-Inference-Benchmarks\" --url \"https://mcp.poorpaul.dev/mcp\""},{"id":"chatgpt","label":"ChatGPT","href":"https://www.mcpbundles.com/docs/integrations/chatgpt"}],"catalogUrl":"https://www.mcpbundles.com/api/directory/v1/servers/poorpaul-mcp-c6ce15878d","callUrl":"https://www.mcpbundles.com/api/directory/v1/servers/poorpaul-mcp-c6ce15878d/tools/{tool}","mcpServerUrl":"https://mcp.poorpaul.dev/mcp","liveCatalogUrl":"https://www.mcpbundles.com/api/directory/v1/servers/poorpaul-mcp-c6ce15878d?live=1"}