--- type: "firecrawl-provider" description: "Artificial Analysis LLM benchmarks: model and reasoning-effort records with Intelligence Index, benchmark scores, cost per task, list prices, output speed and latency, plus per-model and leaderboard-wide API provider endpoint benchmarks." use_when: "Artificial Analysis LLM benchmarks: model and reasoning-effort records with Intelligence Index, benchmark scores, cost per task, list prices, output speed and latency, plus per-model and leaderboard-wide API provider endpoint benchmarks." categories: "AI models" capabilities: 4 credits_per_call: 5 --- # Artificial Analysis on Firecrawl Alexandria Artificial Analysis LLM benchmarks: model and reasoning-effort records with Intelligence Index, benchmark scores, cost per task, list prices, output speed and latency, plus per-model and leaderboard-wide API provider endpoint benchmarks. - Categories: AI models - Category index: [AI models category](https://firecrawl.dev/alexandria/agents/categories/ai-models) - Provider key: `artificialanalysis-ai` - Access: Firecrawl credits - Cost: 5 credits per call ## More - [Human guide](https://firecrawl.dev/app/alexandria/artificialanalysis-ai) - [OpenAPI spec](https://firecrawl.dev/alexandria/agents/providers/artificialanalysis-ai/openapi.json) ## Capabilities - [Get model](https://firecrawl.dev/alexandria/agents/providers/artificialanalysis-ai/benchmarks/get_model): One model record at one reasoning effort from its artificialanalysis.ai page: identity (uuid, slug, release, effort, creator), Intelligence Index with its component evaluations, cost and time per task, benchmark scores, capability indices, list prices, and the first-party provider's output speed, time to first token and end-to-end (500-token) response time with percentiles and per-prompt-type medians. - [List model providers](https://firecrawl.dev/alexandria/agents/providers/artificialanalysis-ai/benchmarks/list_model_providers): Every API provider endpoint Artificial Analysis benchmarks for one model: provider, provider-side model id, list prices, cache pricing, context window and features, Intelligence Index cost and time per task, output speed and time-to-first-token percentiles and per-prompt-type medians. - [Search models](https://firecrawl.dev/alexandria/agents/providers/artificialanalysis-ai/benchmarks/search_models): Search the Artificial Analysis LLM leaderboard: every model and reasoning-effort variant (current and deprecated) with its Intelligence Index, Intelligence Index cost per task (USD, null when unmeasured), benchmark scores, list prices (USD per 1M tokens: the model's list price as the leaderboard states it, not scoped to an API provider; provider-scoped input, output and cache rates are in list_model_providers and search_provider_endpoints), output speed and latency percentiles, sorted by Intelligence Index; reports the index version the page states. Looks up one row by exact `slug`, or filters by text, creator, reasoning and open-weights status; client-side pages over the whole table. - [Search provider endpoints](https://firecrawl.dev/alexandria/agents/providers/artificialanalysis-ai/benchmarks/search_provider_endpoints): Search the Artificial Analysis API providers leaderboard: 1000+ provider endpoints (host x model) with provider-side model id, list prices, cost per Intelligence Index task, context window, function calling / JSON mode support, output speed and time-to-first-token percentiles and the 500-token end-to-end time. Filter by provider, model slug or text; sorted by model Intelligence Index then speed. ## 1. Choose this provider when Artificial Analysis LLM benchmarks: model and reasoning-effort records with Intelligence Index, benchmark scores, cost per task, list prices, output speed and latency, plus per-model and leaderboard-wide API provider endpoint benchmarks. ## 2. Minimal request Call `POST https://api.firecrawl.dev/v2/scrape` with `{ alexandria: { provider, capability, options } }`. For a batch, send `{ alexandria: [...] }` with up to 10 calls. ```json { "provider": "artificialanalysis-ai", "capability": "benchmarks/get_model", "options": { "slug": "claude-fable-5-1" } } ``` ## 3. Add provider options Use only the options needed for the task: - `slug` (string, required): Model slug (e.g. `claude-fable-5-1`, `gpt-6-luna-high`) or a pasted https://artificialanalysis.ai/models/{slug} URL. Pattern: ^[A-Za-z0-9./:_-]+$. Example: `claude-fable-5-1` ## 4. Request through your preferred interface ### JavaScript ```javascript const result = await firecrawl.scrape({ alexandria: { provider: "artificialanalysis-ai", capability: "benchmarks/get_model", options: { slug: "claude-fable-5-1", }, }, }); ``` ### Python ```python result = firecrawl.scrape_alexandria({ "provider": "artificialanalysis-ai", "capability": "benchmarks/get_model", "options": { "slug": "claude-fable-5-1" } }) ``` ### cURL ```sh curl https://api.firecrawl.dev/v2/scrape \ -H "Authorization: Bearer $FIRECRAWL_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "alexandria": { "provider": "artificialanalysis-ai", "capability": "benchmarks/get_model", "options": { "slug": "claude-fable-5-1" } } }' ``` ### CLI ```sh firecrawl scrape 'artificialanalysis-ai/benchmarks/get_model' \ --options '{"slug":"claude-fable-5-1"}' ``` ### MCP Call the FCX MCP retrieve tool with this object: ```json { "provider": "artificialanalysis-ai", "capability": "benchmarks/get_model", "options": { "slug": "claude-fable-5-1" } } ``` Ask for only the returned fields needed by the task. ## 5. Full request shape ```json { "provider": "artificialanalysis-ai", "capability": "benchmarks/get_model", "options": { "slug": "claude-fable-5-1" } } ``` ## 6. Response data The response includes `success`, `provider`, `capability`, `creditsCost` and `data`. This example shows the provider payload in `data`: ```json { "active_parameters_billions": null, "benchmarks": { "aime25": null, "analyst_agent": 0.575, "apex_agents": null, "automation_bench": 0.5937591715646424, "briefcase_elo": 1678.37, "critpt": 0.297142857142857, "gdp_pdf_all_pass": 0.262, "gdpval": 1734.67, "gdpval_normalized": 0.6173350000000001, "gpqa": 0.937373737373737, "harvey_lab": 0.930235924156897, "hle": 0.591288229842447, "ifbench": null, "itbench_sre": null, "lcr": 0.853333333333333, "livecodebench": null, "mlcr_overall": 0.711111111111111, "mmmu_pro": null, "omniscience": 43.45, "omniscience_accuracy": 0.6723333333333333, "omniscience_hallucination_rate": 0.7258392675483215, "scicode": 0.630787037037037, "tau2": null, "tau_banking": 0.472164948453608, "terminal_bench_2_1": 0.913857677902622, "terminal_bench_4_0": 0.52020202020202, "terminal_bench_science": null, "terminalbench_hard": null }, "cache_hit_discount_percent": 0.975, "capabilities": { "economics": 62.6707547111523, "engineering": 56.4696966078099, "finance_and_accounting": 56.416905719465, "healthcare_and_medical": 58.0046680298741, "legal": 60.4916320056283, "strategy_and_ops": 59.6720657328582 }, "context_window_tokens": 1000000, "creator": { "country": "us", "id": "f0aa413f-e8ae-4fcd-9c48-0e049f4f3128", "name": "Anthropic", "slug": "anthropic", "url": "https://www.anthropic.com/" }, "deprecated": false, "deprecated_to": null, "effort": { "label": "max", "level": 60, "slug": "max" }, "host_model_count": 6, "id": "3e87c73e-a257-495e-9730-367a66229811", "intelligence_index": 53.3549259623252, "intelligence_index_cost_per_task_usd": { "answer": 1.543409456346096, "cache_read": 1.603090760194992, "cache_write": 1.8646632226419768, "input": 3.726348513816252, "non_cache_input": 0.25859453097928414, "output": 3.9033578501885873, "reasoning": 2.359948393842491, "total": 7.629706364004841 }, "intelligence_index_evaluations": [ { "cost_per_task_usd": 22.71345865423513, "output_tokens_per_task": 187454.5824175824, "score": 1678.37, "slug": "aa-briefcase", "time_per_task_s": 1736.6822279384282 }, { "cost_per_task_usd": 9.774869276879294, "output_tokens_per_task": 105535.62727272727, "score": 1734.67, "slug": "gdpval-aa", "time_per_task_s": 977.7400260645136 }, { "cost_per_task_usd": 2.611294257588544, "output_tokens_per_task": 31099.32724505327, "score": 0.5937591715646424, "slug": "automationbench-aa", "time_per_task_s": 288.1212517227827 }, { "cost_per_task_usd": 19.21848003106117, "output_tokens_per_task": 191565.7272727273, "score": 0.52020202020202, "slug": "terminalbench-4-0", "time_per_task_s": 1774.7701322954726 }, { "cost_per_task_usd": 0.6585189583333333, "output_tokens_per_task": 12339.597222222224, "score": 0.630787037037037, "slug": "scicode", "time_per_task_s": 114.32080730901193 }, { "cost_per_task_usd": 1.5858915690099744, "output_tokens_per_task": 32010.96061167748, "score": 0.591288229842447, "slug": "humanitys-last-exam", "time_per_task_s": 296.56712402845454 }, { "cost_per_task_usd": 2.7748912999999997, "output_tokens_per_task": 20389.45, "score": 0.262, "slug": "gdp-pdf", "time_per_task_s": 188.89906555369373 }, { "cost_per_task_usd": 5.713794857142857, "output_tokens_per_task": 112959.3, "score": 0.297142857142857, "slug": "critpt", "time_per_task_s": 1046.5170083351616 }, { "cost_per_task_usd": 0.2683618295521425, "output_tokens_per_task": 5371.778333333334, "score": 43.45, "slug": "omniscience", "time_per_task_s": 49.76710541619539 }, { "cost_per_task_usd": 1.5844795862932, "output_tokens_per_task": 3036.39, "score": 0.853333333333333, "slug": "artificial-analysis-long-context-reasoning", "time_per_task_s": 28.130784972452915 } ], "intelligence_index_is_estimated": false, "intelligence_index_output_tokens_per_task": { "answer": 30870.99075157802, "output": 78110.80621282545, "reasoning": 47239.81546124743 }, "intelligence_index_run_cost_usd": { "answer": 2018.4635993445, "cache_read": 1387.5926117156475, "cache_write": 2152.6151098591245, "input": 3891.477294227097, "non_cache_input": 351.269572652325, "output": 9237.381916284525, "reasoning": 7218.918316940026, "total": 13128.859210511622 }, "intelligence_index_time_per_task_s": 723.6614181965862, "is_open_weights": false, "is_reasoning": true, "knowledge_cutoff_date": null, "license_name": null, "license_url": null, "modalities": { "input": [ "text", "image" ], "output": [ "text" ] }, "model_weights_source_url": null, "name": "Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)", "observed_at_ms": 1790202804391, "open_source_categorization": "proprietary", "parameters_billions": null, "performance": { "by_prompt_type": { "hundred_k": { "end_to_end_response_time_s": 99.8404557079705, "output_speed_tokens_per_s": 69.868153076027, "time_to_first_answer_token_s": 92.684119409, "time_to_first_token_s": 92.684119409 }, "long": { "end_to_end_response_time_s": 282.0305711379946, "output_speed_tokens_per_s": 65.8180596366938, "time_to_first_answer_token_s": 274.433871953, "time_to_first_token_s": 274.433871953 }, "medium": { "end_to_end_response_time_s": 121.48977888080192, "output_speed_tokens_per_s": 63.1350703591218, "time_to_first_answer_token_s": 113.5702502075, "time_to_first_token_s": 113.5702502075 }, "medium_parallel": { "end_to_end_response_time_s": 86.81006662192124, "output_speed_tokens_per_s": 78.2662754096591, "time_to_first_answer_token_s": 80.4216190130001, "time_to_first_token_s": 80.4216190130001 } }, "data_source": { "provider_name": "Anthropic", "type": "firstParty" }, "end_to_end_response_time_s": { "answer": 7.596699184994634, "input": 274.433871953, "reasoning": 0, "total": 282.0305711379946 }, "output_speed_tokens_per_s": { "median": 65.8180596366938, "p05": 53.1341320828444, "p95": 76.0462608065299, "q25": 61.7493339506618, "q75": 70.8592592628661 }, "time_to_first_answer_token_s": { "input": 274.433871953, "reasoning": 0, "total": 274.433871953 }, "time_to_first_token_s": { "median": 274.433871953, "p05": 138.2645067669, "p95": 359.8840890979, "q25": 235.058663964, "q75": 326.8302785065 } }, "pricing_usd_per_1m_tokens": { "blended_3_to_1": 20, "blended_7_2_1": 7.175, "cache_hit": 0.25, "cache_write": 12.5, "input": 10, "output": 50 }, "release": { "name": "Claude Fable 5.1", "slug": "claude-fable-5-1" }, "release_date": "2026-09-01", "requested_slug": "claude-fable-5-1", "short_name": "Claude Fable 5.1 (max with fallback)", "size_class": "large", "slug": "claude-fable-5-1", "source_url": "https://artificialanalysis.ai/models/claude-fable-5-1", "url": "https://artificialanalysis.ai/models/claude-fable-5-1" } ``` ## API reference-derived contract The following capability contract is generated from the same normalized Alexandria API reference exposed in the API spec. ### Get model - Capability: `benchmarks/get_model` - Description: One model record at one reasoning effort from its artificialanalysis.ai page: identity (uuid, slug, release, effort, creator), Intelligence Index with its component evaluations, cost and time per task, benchmark scores, capability indices, list prices, and the first-party provider's output speed, time to first token and end-to-end (500-token) response time with percentiles and per-prompt-type medians. - Instructions: When you have a model slug from search_models (or a pasted /models/{slug} URL) and need the full benchmark record. - Cost: 5 credits per call - Capability file: [Get model](https://firecrawl.dev/alexandria/agents/providers/artificialanalysis-ai/benchmarks/get_model) Accepted options: - `slug` (string, required): Model slug (e.g. `claude-fable-5-1`, `gpt-6-luna-high`) or a pasted https://artificialanalysis.ai/models/{slug} URL. Pattern: ^[A-Za-z0-9./:_-]+$. Example: `claude-fable-5-1` Response schema example: ```json { "active_parameters_billions": null, "benchmarks": { "aime25": null, "analyst_agent": 0.575, "apex_agents": null, "automation_bench": 0.5937591715646424, "briefcase_elo": 1678.37, "critpt": 0.297142857142857, "gdp_pdf_all_pass": 0.262, "gdpval": 1734.67, "gdpval_normalized": 0.6173350000000001, "gpqa": 0.937373737373737, "harvey_lab": 0.930235924156897, "hle": 0.591288229842447, "ifbench": null, "itbench_sre": null, "lcr": 0.853333333333333, "livecodebench": null, "mlcr_overall": 0.711111111111111, "mmmu_pro": null, "omniscience": 43.45, "omniscience_accuracy": 0.6723333333333333, "omniscience_hallucination_rate": 0.7258392675483215, "scicode": 0.630787037037037, "tau2": null, "tau_banking": 0.472164948453608, "terminal_bench_2_1": 0.913857677902622, "terminal_bench_4_0": 0.52020202020202, "terminal_bench_science": null, "terminalbench_hard": null }, "cache_hit_discount_percent": 0.975, "capabilities": { "economics": 62.6707547111523, "engineering": 56.4696966078099, "finance_and_accounting": 56.416905719465, "healthcare_and_medical": 58.0046680298741, "legal": 60.4916320056283, "strategy_and_ops": 59.6720657328582 }, "context_window_tokens": 1000000, "creator": { "country": "us", "id": "f0aa413f-e8ae-4fcd-9c48-0e049f4f3128", "name": "Anthropic", "slug": "anthropic", "url": "https://www.anthropic.com/" }, "deprecated": false, "deprecated_to": null, "effort": { "label": "max", "level": 60, "slug": "max" }, "host_model_count": 6, "id": "3e87c73e-a257-495e-9730-367a66229811", "intelligence_index": 53.3549259623252, "intelligence_index_cost_per_task_usd": { "answer": 1.543409456346096, "cache_read": 1.603090760194992, "cache_write": 1.8646632226419768, "input": 3.726348513816252, "non_cache_input": 0.25859453097928414, "output": 3.9033578501885873, "reasoning": 2.359948393842491, "total": 7.629706364004841 }, "intelligence_index_evaluations": [ { "cost_per_task_usd": 22.71345865423513, "output_tokens_per_task": 187454.5824175824, "score": 1678.37, "slug": "aa-briefcase", "time_per_task_s": 1736.6822279384282 }, { "cost_per_task_usd": 9.774869276879294, "output_tokens_per_task": 105535.62727272727, "score": 1734.67, "slug": "gdpval-aa", "time_per_task_s": 977.7400260645136 }, { "cost_per_task_usd": 2.611294257588544, "output_tokens_per_task": 31099.32724505327, "score": 0.5937591715646424, "slug": "automationbench-aa", "time_per_task_s": 288.1212517227827 }, { "cost_per_task_usd": 19.21848003106117, "output_tokens_per_task": 191565.7272727273, "score": 0.52020202020202, "slug": "terminalbench-4-0", "time_per_task_s": 1774.7701322954726 }, { "cost_per_task_usd": 0.6585189583333333, "output_tokens_per_task": 12339.597222222224, "score": 0.630787037037037, "slug": "scicode", "time_per_task_s": 114.32080730901193 }, { "cost_per_task_usd": 1.5858915690099744, "output_tokens_per_task": 32010.96061167748, "score": 0.591288229842447, "slug": "humanitys-last-exam", "time_per_task_s": 296.56712402845454 }, { "cost_per_task_usd": 2.7748912999999997, "output_tokens_per_task": 20389.45, "score": 0.262, "slug": "gdp-pdf", "time_per_task_s": 188.89906555369373 }, { "cost_per_task_usd": 5.713794857142857, "output_tokens_per_task": 112959.3, "score": 0.297142857142857, "slug": "critpt", "time_per_task_s": 1046.5170083351616 }, { "cost_per_task_usd": 0.2683618295521425, "output_tokens_per_task": 5371.778333333334, "score": 43.45, "slug": "omniscience", "time_per_task_s": 49.76710541619539 }, { "cost_per_task_usd": 1.5844795862932, "output_tokens_per_task": 3036.39, "score": 0.853333333333333, "slug": "artificial-analysis-long-context-reasoning", "time_per_task_s": 28.130784972452915 } ], "intelligence_index_is_estimated": false, "intelligence_index_output_tokens_per_task": { "answer": 30870.99075157802, "output": 78110.80621282545, "reasoning": 47239.81546124743 }, "intelligence_index_run_cost_usd": { "answer": 2018.4635993445, "cache_read": 1387.5926117156475, "cache_write": 2152.6151098591245, "input": 3891.477294227097, "non_cache_input": 351.269572652325, "output": 9237.381916284525, "reasoning": 7218.918316940026, "total": 13128.859210511622 }, "intelligence_index_time_per_task_s": 723.6614181965862, "is_open_weights": false, "is_reasoning": true, "knowledge_cutoff_date": null, "license_name": null, "license_url": null, "modalities": { "input": [ "text", "image" ], "output": [ "text" ] }, "model_weights_source_url": null, "name": "Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)", "observed_at_ms": 1790202804391, "open_source_categorization": "proprietary", "parameters_billions": null, "performance": { "by_prompt_type": { "hundred_k": { "end_to_end_response_time_s": 99.8404557079705, "output_speed_tokens_per_s": 69.868153076027, "time_to_first_answer_token_s": 92.684119409, "time_to_first_token_s": 92.684119409 }, "long": { "end_to_end_response_time_s": 282.0305711379946, "output_speed_tokens_per_s": 65.8180596366938, "time_to_first_answer_token_s": 274.433871953, "time_to_first_token_s": 274.433871953 }, "medium": { "end_to_end_response_time_s": 121.48977888080192, "output_speed_tokens_per_s": 63.1350703591218, "time_to_first_answer_token_s": 113.5702502075, "time_to_first_token_s": 113.5702502075 }, "medium_parallel": { "end_to_end_response_time_s": 86.81006662192124, "output_speed_tokens_per_s": 78.2662754096591, "time_to_first_answer_token_s": 80.4216190130001, "time_to_first_token_s": 80.4216190130001 } }, "data_source": { "provider_name": "Anthropic", "type": "firstParty" }, "end_to_end_response_time_s": { "answer": 7.596699184994634, "input": 274.433871953, "reasoning": 0, "total": 282.0305711379946 }, "output_speed_tokens_per_s": { "median": 65.8180596366938, "p05": 53.1341320828444, "p95": 76.0462608065299, "q25": 61.7493339506618, "q75": 70.8592592628661 }, "time_to_first_answer_token_s": { "input": 274.433871953, "reasoning": 0, "total": 274.433871953 }, "time_to_first_token_s": { "median": 274.433871953, "p05": 138.2645067669, "p95": 359.8840890979, "q25": 235.058663964, "q75": 326.8302785065 } }, "pricing_usd_per_1m_tokens": { "blended_3_to_1": 20, "blended_7_2_1": 7.175, "cache_hit": 0.25, "cache_write": 12.5, "input": 10, "output": 50 }, "release": { "name": "Claude Fable 5.1", "slug": "claude-fable-5-1" }, "release_date": "2026-09-01", "requested_slug": "claude-fable-5-1", "short_name": "Claude Fable 5.1 (max with fallback)", "size_class": "large", "slug": "claude-fable-5-1", "source_url": "https://artificialanalysis.ai/models/claude-fable-5-1", "url": "https://artificialanalysis.ai/models/claude-fable-5-1" } ``` ### List model providers - Capability: `benchmarks/list_model_providers` - Description: Every API provider endpoint Artificial Analysis benchmarks for one model: provider, provider-side model id, list prices, cache pricing, context window and features, Intelligence Index cost and time per task, output speed and time-to-first-token percentiles and per-prompt-type medians. - Instructions: To compare where to run a given model: prices and measured speed per provider. Needs a model slug from search_models. - Cost: 5 credits per call - Capability file: [List model providers](https://firecrawl.dev/alexandria/agents/providers/artificialanalysis-ai/benchmarks/list_model_providers) Accepted options: - `slug` (string, required): Model slug (e.g. `gpt-6-astra`) or a pasted https://artificialanalysis.ai/models/{slug}/providers URL. Pattern: ^[A-Za-z0-9./:_-]+$. Example: `gpt-6-astra` Response schema example: ```json { "count": 1, "endpoints": [ { "features": { "context_window_tokens": 1000000, "function_calling": true, "json_mode": false, "openai_compatible": true }, "host": { "label": "Anthropic", "name": "Anthropic", "slug": "anthropic" }, "host_api_id": "claude-fable-5-1", "id": "c7e37b8e-9e3c-4bc2-a055-1f6db77cdad1", "intelligence_index_cost_per_task_usd": { "answer": 1.543409456346096, "cache_read": 1.603090760194992, "cache_write": 1.8646632226419768, "input": 3.726348513816252, "non_cache_input": 0.25859453097928414, "output": 3.9033578501885873, "reasoning": 2.359948393842491, "total": 7.629706364004841 }, "intelligence_index_time_per_task_s": 723.6614181965862, "label": "Claude Fable 5.1 (max with fallback)", "model_slug": "claude-fable-5-1", "performance": { "by_prompt_type": { "hundred_k": { "end_to_end_response_time_s": 99.8404557079705, "output_speed_tokens_per_s": 69.868153076027, "time_to_first_answer_token_s": null, "time_to_first_token_s": 92.684119409 }, "long": { "end_to_end_response_time_s": 282.0305711379946, "output_speed_tokens_per_s": 65.8180596366938, "time_to_first_answer_token_s": null, "time_to_first_token_s": 274.433871953 }, "medium": { "end_to_end_response_time_s": 121.48977888080192, "output_speed_tokens_per_s": 63.1350703591218, "time_to_first_answer_token_s": null, "time_to_first_token_s": 113.5702502075 }, "medium_parallel": { "end_to_end_response_time_s": 86.81006662192124, "output_speed_tokens_per_s": 78.2662754096591, "time_to_first_answer_token_s": null, "time_to_first_token_s": 80.4216190130001 }, "vision_single_image": { "end_to_end_response_time_s": 33.56031834653706, "output_speed_tokens_per_s": 39.059959642339, "time_to_first_answer_token_s": null, "time_to_first_token_s": 20.759485868 } }, "end_to_end_response_time_s": { "answer": 7.596699184994634, "input": 274.433871953, "reasoning": 0, "total": 282.0305711379946 }, "output_speed_tokens_per_s": { "median": 65.8180596366938, "p05": 53.1341320828444, "p95": 76.0462608065299, "q25": 61.7493339506618, "q75": 70.8592592628661 }, "time_to_first_answer_token_s": { "input": 274.433871953, "reasoning": 0, "total": 274.433871953 }, "time_to_first_token_s": { "median": 274.433871953, "p05": 138.2645067669, "p95": 359.8840890979, "q25": 235.058663964, "q75": 326.8302785065 } }, "price_class": "high", "pricing_usd_per_1m_tokens": { "blended_3_to_1": 20, "blended_7_2_1": 7.175, "cache_hit": 0.25, "cache_hit_discount_percent": 0.975, "cache_hit_rate": 0.994524976, "cache_write": 12.5, "input": 10, "output": 50 }, "slug": "anthropic_claude-fable-5-1", "url": "https://artificialanalysis.ai/models/claude-fable-5-1/providers" } ], "model_slug": "claude-fable-5-1", "observed_at_ms": 1790202820479, "source_url": "https://artificialanalysis.ai/models/claude-fable-5-1/providers" } ``` ### Search models - Capability: `benchmarks/search_models` - Description: Search the Artificial Analysis LLM leaderboard: every model and reasoning-effort variant (current and deprecated) with its Intelligence Index, Intelligence Index cost per task (USD, null when unmeasured), benchmark scores, list prices (USD per 1M tokens: the model's list price as the leaderboard states it, not scoped to an API provider; provider-scoped input, output and cache rates are in list_model_providers and search_provider_endpoints), output speed and latency percentiles, sorted by Intelligence Index; reports the index version the page states. Looks up one row by exact `slug`, or filters by text, creator, reasoning and open-weights status; client-side pages over the whole table. - Instructions: Start here to find a model slug (e.g. `claude-fable-5-1-high`) for get_model or list_model_providers, to rank models by intelligence, price or speed, or with `slug` to fetch one known model/effort row (legacy slugs included) without paging the whole table. - Cost: 5 credits per call - Capability file: [Search models](https://firecrawl.dev/alexandria/agents/providers/artificialanalysis-ai/benchmarks/search_models) Accepted options: - `creator` (string): Creator name filter, case-insensitive substring (e.g. `anthropic`, `openai`). Example: `` - `cursor` (string): Opaque `next_cursor` from the previous page. Pattern: ^[0-9]+$. Example: `next_cursor` - `include_deprecated` (boolean): Include models the site marks deprecated (most of the table: 420 of 684 rows on 2026-09-29). Implied true when `slug` is given. Example: `false` - `limit` (number): Results per page (1-100). Example: `20` - `open_weights` (boolean): Only open-weights (true) or proprietary (false) models. Example: `false` - `query` (string): Case-insensitive terms matched against model name, short name, slug and creator; every whitespace-separated term must match (substring: `deepseek-v3` also matches `deepseek-v3-0324`; use `slug` for an exact row). Punctuation is folded on both sides, so `Claude Sonnet 5.5`, `claude-sonnet-5.5` and `claude-sonnet-5-5` find the same rows. One model per call: a list of several names matches nothing. Example: `deepseek-v3` - `reasoning` (boolean): Only reasoning (true) or non-reasoning (false) models. Example: `false` - `slug` (string): Exact model slug at one reasoning effort (e.g. `claude-fable-5-1-high`, `deepseek-v3`) or a pasted https://artificialanalysis.ai/models/{slug} URL: returns that one row (effort siblings such as `-high`/`-low` are not matched). Implies `include_deprecated: true` so legacy slugs resolve; the row's `deprecated` says whether the site still lists it. Combines with the other filters. Pattern: ^[A-Za-z0-9./:_-]+$. Example: `claude-fable-5-1-high` Response schema example: ```json { "count": 5, "intelligence_index_version": "4.3", "next_cursor": null, "observed_at_ms": 1790708711649, "query": { "creator": null, "cursor": null, "include_deprecated": false, "limit": 5, "open_weights": null, "query": "claude fable", "reasoning": null, "slug": null }, "results": [ { "benchmarks": { "analyst_agent": 0.575, "apex_agents": null, "critpt": 0.297142857142857, "gdpval_normalized": 0.6173350000000001, "gpqa": 0.937373737373737, "hle": 0.591288229842447, "ifbench": null, "itbench_sre": 0.495291902071563, "lcr": 0.853333333333333, "mmmu_pro": null, "omniscience": 43.45, "omniscience_accuracy": 0.6723333333333333, "omniscience_non_hallucination": 0.2741607324516785, "scicode": 0.630787037037037, "tau2": null, "tau_banking": 0.472164948453608, "terminal_bench_2_1": 0.913857677902622, "terminal_bench_4_0": 0.52020202020202, "terminalbench_hard": null }, "context_window_tokens": 1000000, "creator_name": "Anthropic", "deprecated": false, "intelligence_index": 53.3549259623252, "intelligence_index_cost_per_task_usd": 7.629706364004841, "intelligence_index_is_estimated": false, "is_open_weights": false, "is_reasoning": true, "name": "Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)", "performance": { "end_to_end_response_time_s": 293.20013133105056, "output_speed_tokens_per_s": { "median": 69.2781577129234, "p05": 59.4408879866233, "p95": 78.9948972499949, "q25": 62.8090791995804, "q75": 71.2729428399078 }, "reasoning_time_s": null, "time_to_first_answer_token_s": 285.9828493405, "time_to_first_token_s": { "median": 285.9828493405, "p05": 61.2510733455, "p95": 353.80174481275, "q25": 186.0576103785, "q75": 325.142944658 } }, "price_class": "high", "pricing_usd_per_1m_tokens": { "cache_hit": 0.25, "cache_write": 12.5, "input": 10, "output": 50 }, "short_name": "Claude Fable 5.1 (max with fallback)", "size_class": null, "slug": "claude-fable-5-1", "url": "https://artificialanalysis.ai/models/claude-fable-5-1" }, { "benchmarks": { "analyst_agent": null, "apex_agents": null, "critpt": 0.311428571428571, "gdpval_normalized": 0.61046, "gpqa": 0.934343434343434, "hle": 0.587117701575533, "ifbench": null, "itbench_sre": null, "lcr": 0.83, "mmmu_pro": null, "omniscience": 42.38333333333333, "omniscience_accuracy": 0.6621666666666667, "omniscience_non_hallucination": 0.2945239269856932, "scicode": 0.608796296296296, "tau2": null, "tau_banking": 0.457731958762887, "terminal_bench_2_1": 0.910112359550562, "terminal_bench_4_0": 0.55050505050505, "terminalbench_hard": null }, "context_window_tokens": 1000000, "creator_name": "Anthropic", "deprecated": false, "intelligence_index": 53.2032587538944, "intelligence_index_cost_per_task_usd": 5.978304247388257, "intelligence_index_is_estimated": false, "is_open_weights": false, "is_reasoning": true, "name": "Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)", "performance": { "end_to_end_response_time_s": 113.29741511822373, "output_speed_tokens_per_s": { "median": 56.8889512971994, "p05": 47.2736525057793, "p95": 73.0776195059553, "q25": 52.6646444921629, "q75": 67.9315481066607 }, "reasoning_time_s": null, "time_to_first_answer_token_s": 104.50836226, "time_to_first_token_s": { "median": 104.50836226, "p05": 36.944357231, "p95": 248.6625857076, "q25": 57.014412537, "q75": 179.860018353 } }, "price_class": "high", "pricing_usd_per_1m_tokens": { "cache_hit": 0.25, "cache_write": 12.5, "input": 10, "output": 50 }, "short_name": "Claude Fable 5.1 (xhigh with fallback)", "size_class": null, "slug": "claude-fable-5-1-xhigh", "url": "https://artificialanalysis.ai/models/claude-fable-5-1-xhigh" }, { "benchmarks": { "analyst_agent": null, "apex_agents": null, "critpt": 0.302857142857143, "gdpval_normalized": 0.558645, "gpqa": 0.906060606060606, "hle": 0.559314179796108, "ifbench": null, "itbench_sre": null, "lcr": 0.836666666666667, "mmmu_pro": null, "omniscience": 40.8, "omniscience_accuracy": 0.6493333333333333, "omniscience_non_hallucination": 0.311787072243346, "scicode": 0.586805555555556, "tau2": null, "tau_banking": 0.430927835051546, "terminal_bench_2_1": 0.898876404494382, "terminal_bench_4_0": 0.52020202020202, "terminalbench_hard": null }, "context_window_tokens": 1000000, "creator_name": "Anthropic", "deprecated": false, "intelligence_index": 51.1515708852237, "intelligence_index_cost_per_task_usd": 3.912526554325558, "intelligence_index_is_estimated": false, "is_open_weights": false, "is_reasoning": true, "name": "Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)", "performance": { "end_to_end_response_time_s": 35.58863103214977, "output_speed_tokens_per_s": { "median": 52.2121460921833, "p05": 45.0985260425514, "p95": 69.6803080027435, "q25": 47.1217196910813, "q75": 62.4946380395491 }, "reasoning_time_s": null, "time_to_first_answer_token_s": 26.012315224, "time_to_first_token_s": { "median": 26.012315224, "p05": 4.93500340379989, "p95": 44.0248097684999, "q25": 12.1694669465, "q75": 31.3693713115 } }, "price_class": "high", "pricing_usd_per_1m_tokens": { "cache_hit": 0.25, "cache_write": 12.5, "input": 10, "output": 50 }, "short_name": "Claude Fable 5.1 (high with fallback)", "size_class": null, "slug": "claude-fable-5-1-high", "url": "https://artificialanalysis.ai/models/claude-fable-5-1-high" }, { "benchmarks": { "analyst_agent": null, "apex_agents": null, "critpt": 0.291428571428571, "gdpval_normalized": 0.517925, "gpqa": 0.885858585858586, "hle": 0.537998146431881, "ifbench": null, "itbench_sre": null, "lcr": 0.846666666666667, "mmmu_pro": null, "omniscience": 37.6, "omniscience_accuracy": 0.631, "omniscience_non_hallucination": 0.3089430894308943, "scicode": 0.563657407407407, "tau2": null, "tau_banking": 0.410309278350515, "terminal_bench_2_1": 0.880149812734082, "terminal_bench_4_0": 0.44949494949495, "terminalbench_hard": null }, "context_window_tokens": 1000000, "creator_name": "Anthropic", "deprecated": false, "intelligence_index": 48.9206162609523, "intelligence_index_cost_per_task_usd": 2.982597436096585, "intelligence_index_is_estimated": false, "is_open_weights": false, "is_reasoning": true, "name": "Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback)", "performance": { "end_to_end_response_time_s": 17.90313303398304, "output_speed_tokens_per_s": { "median": 50.4859653874342, "p05": 43.0005067700101, "p95": 75.6108191622584, "q25": 46.8762444036515, "q75": 61.2663377555063 }, "reasoning_time_s": null, "time_to_first_answer_token_s": 7.99939055499999, "time_to_first_token_s": { "median": 7.99939055499999, "p05": 3.4419402559, "p95": 35.2594710314001, "q25": 5.59770872150011, "q75": 18.9534529089999 } }, "price_class": "high", "pricing_usd_per_1m_tokens": { "cache_hit": 0.25, "cache_write": 12.5, "input": 10, "output": 50 }, "short_name": "Claude Fable 5.1 (medium with fallback)", "size_class": null, "slug": "claude-fable-5-1-medium", "url": "https://artificialanalysis.ai/models/claude-fable-5-1-medium" }, { "benchmarks": { "analyst_agent": null, "apex_agents": null, "critpt": 0.277142857142857, "gdpval_normalized": 0.474785, "gpqa": 0.880808080808081, "hle": 0.48887859128823, "ifbench": null, "itbench_sre": null, "lcr": 0.823333333333333, "mmmu_pro": null, "omniscience": 34.13333333333333, "omniscience_accuracy": 0.6023333333333334, "omniscience_non_hallucination": 0.34367141659681477, "scicode": 0.56712962962963, "tau2": null, "tau_banking": 0.389690721649485, "terminal_bench_2_1": 0.850187265917603, "terminal_bench_4_0": 0.404040404040404, "terminalbench_hard": null }, "context_window_tokens": 1000000, "creator_name": "Anthropic", "deprecated": false, "intelligence_index": 46.8163422014862, "intelligence_index_cost_per_task_usd": 2.371025320183763, "intelligence_index_is_estimated": false, "is_open_weights": false, "is_reasoning": true, "name": "Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback)", "performance": { "end_to_end_response_time_s": 15.276830952412558, "output_speed_tokens_per_s": { "median": 47.876953813804, "p05": 41.6449573256389, "p95": 70.5310768146966, "q25": 45.5685679814564, "q75": 60.2368065300317 }, "reasoning_time_s": null, "time_to_first_answer_token_s": 4.83339292700003, "time_to_first_token_s": { "median": 4.83339292700003, "p05": 2.16069615300003, "p95": 8.20163022479991, "q25": 3.9688941710001, "q75": 5.45523965999996 } }, "price_class": "high", "pricing_usd_per_1m_tokens": { "cache_hit": 0.25, "cache_write": 12.5, "input": 10, "output": 50 }, "short_name": "Claude Fable 5.1 (low with fallback)", "size_class": null, "slug": "claude-fable-5-1-low", "url": "https://artificialanalysis.ai/models/claude-fable-5-1-low" } ], "source_url": "https://artificialanalysis.ai/leaderboards/models", "table_size": 684, "total_matches": 5 } ``` ### Search provider endpoints - Capability: `benchmarks/search_provider_endpoints` - Description: Search the Artificial Analysis API providers leaderboard: 1000+ provider endpoints (host x model) with provider-side model id, list prices, cost per Intelligence Index task, context window, function calling / JSON mode support, output speed and time-to-first-token percentiles and the 500-token end-to-end time. Filter by provider, model slug or text; sorted by model Intelligence Index then speed. - Instructions: To find the fastest or cheapest endpoint for a model, or all models a provider hosts. Use list_model_providers for one model's percentile detail. - Cost: 5 credits per call - Capability file: [Search provider endpoints](https://firecrawl.dev/alexandria/agents/providers/artificialanalysis-ai/benchmarks/search_provider_endpoints) Accepted options: - `cursor` (string): Opaque `next_cursor` from the previous page. Pattern: ^[0-9]+$. Example: `next_cursor` - `host` (string): API provider slug or name, exact and case-insensitive (e.g. `anthropic`, `amazon_bedrock`, `novita`). Example: `` - `include_deprecated` (boolean): Include endpoints of deprecated models. Example: `false` - `limit` (number): Results per page (1-100). Example: `20` - `model` (string): Model slug, exact (e.g. `claude-fable-5-1`). Pattern: ^[A-Za-z0-9./:_-]+$. Example: `claude-fable-5-1` - `query` (string): Case-insensitive terms matched against endpoint label, model slug and provider; every term must match. Example: `` Response schema example: ```json { "count": 1, "next_cursor": null, "observed_at_ms": 1790202950957, "query": { "cursor": null, "host": "anthropic", "include_deprecated": false, "limit": 20, "model": "claude-fable-5-1", "query": null }, "results": [ { "features": { "context_window_tokens": 1000000, "function_calling": true, "json_mode": false, "openai_compatible": true }, "host": { "function_calling_url": "https://docs.anthropic.com/en/docs/build-with-claude/tool-use", "json_mode_url": "https://docs.anthropic.com/en/docs/build-with-claude/tool-use#json-mode", "name": "Anthropic", "slug": "anthropic" }, "host_api_id": "claude-fable-5-1", "id": "c7e37b8e-9e3c-4bc2-a055-1f6db77cdad1", "intelligence_index_cost_per_task_usd": 7.629706364004841, "label": "Claude Fable 5.1 (max with fallback)", "model": { "benchmarks": { "aime25": null, "apex_agents": null, "automation_bench": 0.5937591715646424, "briefcase_elo": 1678.37, "critpt": 0.297142857142857, "gdpval_normalized": 0.6173350000000001, "gpqa": 0.937373737373737, "harvey_lab": 0.930235924156897, "hle": 0.591288229842447, "ifbench": null, "itbench_sre": null, "lcr": 0.853333333333333, "livecodebench": null, "mmmu_pro": null, "omniscience": 43.45, "omniscience_accuracy": 0.6723333333333333, "omniscience_non_hallucination": 0.2741607324516785, "scicode": 0.630787037037037, "tau2": null, "tau_banking": 0.472164948453608, "terminal_bench_2_1": 0.913857677902622, "terminal_bench_4_0": 0.52020202020202, "terminalbench_hard": null }, "creator_name": "Anthropic", "deprecated": false, "intelligence_index": 53.3549259623252, "intelligence_index_is_estimated": false, "is_open_weights": false, "is_reasoning": true, "size_class": "large", "slug": "claude-fable-5-1", "url": "https://artificialanalysis.ai/models/claude-fable-5-1" }, "performance": { "end_to_end_response_time_s": 282.0305711379946, "output_speed_tokens_per_s": { "median": 65.8180596366938, "p05": 53.1341320828444, "p95": 76.0462608065299, "q25": 61.7493339506618, "q75": 70.8592592628661 }, "reasoning_time_s": null, "time_to_first_answer_token_s": 274.433871953, "time_to_first_token_s": { "median": 274.433871953, "p05": 138.2645067669, "p95": 359.8840890979, "q25": 235.058663964, "q75": 326.8302785065 } }, "price_class": "high", "pricing_notes": null, "pricing_usd_per_1m_tokens": { "cache_hit": 0.25, "cache_write": 12.5, "input": 10, "output": 50 }, "url": "https://artificialanalysis.ai/models/claude-fable-5-1/providers" } ], "source_url": "https://artificialanalysis.ai/leaderboards/providers", "table_size": 1189, "total_matches": 1 } ```