{"schemaVersion":1,"revision":"efe81a7a0d70ac85762b9e43f63acffae74f123554667232b55cec5e6db10663","verifiedAt":"2026-09-22T00:00:00Z","models":[{"provider":"anthropic","id":"claude-fable-5-1","label":"Claude Fable 5.1","selection":"default","contextWindow":1000000,"maxOutput":128000,"thinkingMode":"adaptive","skill":10,"description":"Hardest work: architecture, multi-service changes, long-context debugging, anything where a wrong call is expensive.","source":{"urls":["https://platform.claude.com/docs/en/models/fable-5-1/overview","https://platform.claude.com/docs/en/build-with-claude/vision","https://platform.claude.com/docs/en/about-claude/pricing"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"reasoning":true,"imageInput":{"supported":true,"mimeTypes":["image/jpeg","image/png","image/gif","image/webp"],"maxImages":600,"maxWidth":8000,"maxHeight":8000,"notes":["Direct Claude API: 10 MB base64-encoded per image; 32 MB whole request. MB byte definition unspecified.","Animation uses the first frame. More than 20 images has stricter dimensions; Pumpkin caps requests at 20 images."],"sourceUrls":["https://platform.claude.com/docs/en/models/fable-5-1/overview","https://platform.claude.com/docs/en/build-with-claude/vision"]}},"clientPolicy":{"images":{"maxBytesPerImage":7000000,"maxLongEdge":2576,"maxPixels":3750656,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":20000000,"maxImages":20,"outputMimeType":"image/png","patchSize":28,"maxPatches":4784}},"pricing":{"input":10,"output":50,"cacheRead":0.25,"cacheWrite":12.5,"webSearchPerCall":0.01},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://platform.claude.com/docs/en/about-claude/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["cacheWrite is the 5-minute TTL rate. One-hour cache write: $20 per million tokens. Cache writes replace ordinary input charges, not an additive fee.","Standard global routing; Batch, fast modes and regional processing differ."]},"notes":["Standard Messages output limit; thinking shares output budget.","Adaptive thinking is always on; forced tool choice is unsupported."]},{"provider":"anthropic","id":"claude-opus-5-5","label":"Claude Opus 5.5","selection":"default","contextWindow":1000000,"maxOutput":128000,"thinkingMode":"adaptive","skill":9,"description":"Long-running agentic coding and knowledge work. Newer and cheaper than Opus 5, still below Fable.","source":{"urls":["https://platform.claude.com/docs/en/models/opus-5-5/overview","https://platform.claude.com/docs/en/build-with-claude/vision","https://platform.claude.com/docs/en/about-claude/pricing"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"reasoning":true,"imageInput":{"supported":true,"mimeTypes":["image/jpeg","image/png","image/gif","image/webp"],"maxImages":600,"maxWidth":8000,"maxHeight":8000,"notes":["Direct Claude API: 10 MB base64-encoded per image; 32 MB whole request. MB byte definition unspecified.","Animation uses the first frame. More than 20 images has stricter dimensions; Pumpkin caps requests at 20 images."],"sourceUrls":["https://platform.claude.com/docs/en/models/opus-5-5/overview","https://platform.claude.com/docs/en/build-with-claude/vision"]}},"clientPolicy":{"images":{"maxBytesPerImage":7000000,"maxLongEdge":2576,"maxPixels":3750656,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":20000000,"maxImages":20,"outputMimeType":"image/png","patchSize":28,"maxPatches":4784}},"pricing":{"input":4,"output":20,"cacheRead":0.2,"cacheWrite":5,"webSearchPerCall":0.01},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://platform.claude.com/docs/en/about-claude/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["cacheWrite is the 5-minute TTL rate. One-hour cache write: $8 per million tokens. Cache writes replace ordinary input charges, not an additive fee.","Cache read is $0.20 per million tokens (0.05x base input).","Standard global routing; Batch, fast modes and regional processing differ."]},"notes":["Standard Messages output limit; thinking shares output budget.","Adaptive thinking is always on; forced tool choice is unsupported. Default effort is medium.","Cache read is 0.05x base input, not the usual 0.1x."]},{"provider":"anthropic","id":"claude-opus-5","label":"Claude Opus 5","selection":"default","contextWindow":1000000,"maxOutput":128000,"thinkingMode":"adaptive","skill":8,"description":"Strong implementation and review when Fable is overkill: scoped features, correctness-sensitive edits, production-ready diffs.","source":{"urls":["https://platform.claude.com/docs/en/models/opus-5/overview","https://platform.claude.com/docs/en/build-with-claude/vision","https://platform.claude.com/docs/en/about-claude/pricing"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"reasoning":true,"imageInput":{"supported":true,"mimeTypes":["image/jpeg","image/png","image/gif","image/webp"],"maxImages":600,"maxWidth":8000,"maxHeight":8000,"notes":["Direct Claude API: 10 MB base64-encoded per image; 32 MB whole request. MB byte definition unspecified.","Animation uses the first frame. More than 20 images has stricter dimensions; Pumpkin caps requests at 20 images."],"sourceUrls":["https://platform.claude.com/docs/en/models/opus-5/overview","https://platform.claude.com/docs/en/build-with-claude/vision"]}},"clientPolicy":{"images":{"maxBytesPerImage":7000000,"maxLongEdge":2576,"maxPixels":3750656,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":20000000,"maxImages":20,"outputMimeType":"image/png","patchSize":28,"maxPatches":4784}},"pricing":{"input":5,"output":25,"cacheRead":0.5,"cacheWrite":6.25,"webSearchPerCall":0.01},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://platform.claude.com/docs/en/about-claude/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["cacheWrite is the 5-minute TTL rate. One-hour cache write: $10 per million tokens. Cache writes replace ordinary input charges, not an additive fee.","Standard global routing; Batch, fast modes and regional processing differ."]},"notes":["Standard Messages output limit; thinking shares output budget."]},{"provider":"anthropic","id":"claude-sonnet-5","label":"Claude Sonnet 5","selection":"default","contextWindow":1000000,"maxOutput":128000,"thinkingMode":"adaptive","skill":6,"description":"Everyday coding: routine features, refactors with a known shape, tests, docs. Mid-tier default for Anthropic.","source":{"urls":["https://platform.claude.com/docs/en/models/sonnet-5/overview","https://platform.claude.com/docs/en/build-with-claude/vision","https://platform.claude.com/docs/en/about-claude/pricing"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"reasoning":true,"imageInput":{"supported":true,"mimeTypes":["image/jpeg","image/png","image/gif","image/webp"],"maxImages":600,"maxWidth":8000,"maxHeight":8000,"notes":["Direct Claude API: 10 MB base64-encoded per image; 32 MB whole request. MB byte definition unspecified.","Animation uses the first frame. More than 20 images has stricter dimensions; Pumpkin caps requests at 20 images."],"sourceUrls":["https://platform.claude.com/docs/en/models/sonnet-5/overview","https://platform.claude.com/docs/en/build-with-claude/vision"]}},"clientPolicy":{"images":{"maxBytesPerImage":7000000,"maxLongEdge":2576,"maxPixels":3750656,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":20000000,"maxImages":20,"outputMimeType":"image/png","patchSize":28,"maxPatches":4784}},"pricing":{"input":2,"output":10,"cacheRead":0.2,"cacheWrite":2.5,"webSearchPerCall":0.01},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://platform.claude.com/docs/en/about-claude/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["cacheWrite is the 5-minute TTL rate. One-hour cache write: $4 per million tokens. Cache writes replace ordinary input charges, not an additive fee.","Standard global routing; Batch, fast modes and regional processing differ."]},"notes":["Standard Messages output limit; thinking shares output budget."]},{"provider":"anthropic","id":"claude-haiku-4-5","label":"Claude Haiku 4.5","selection":"default","contextWindow":200000,"maxOutput":64000,"thinkingMode":"extended","skill":4,"description":"Cheap, short jobs: mechanical edits, triage, titles, compaction, apply this exact change. Not for design or long hunts.","source":{"urls":["https://platform.claude.com/docs/en/models/haiku-4-5/overview","https://platform.claude.com/docs/en/build-with-claude/vision","https://platform.claude.com/docs/en/about-claude/pricing"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"reasoning":true,"imageInput":{"supported":true,"mimeTypes":["image/jpeg","image/png","image/gif","image/webp"],"maxImages":100,"maxWidth":8000,"maxHeight":8000,"notes":["Direct Claude API: 10 MB base64-encoded per image; 32 MB whole request. MB byte definition unspecified.","Animation uses the first frame. More than 20 images has stricter dimensions; Pumpkin caps requests at 20 images."],"sourceUrls":["https://platform.claude.com/docs/en/models/haiku-4-5/overview","https://platform.claude.com/docs/en/build-with-claude/vision"]}},"clientPolicy":{"images":{"maxBytesPerImage":7000000,"maxLongEdge":1568,"maxPixels":1229312,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":20000000,"maxImages":20,"outputMimeType":"image/png","patchSize":28,"maxPatches":1568}},"pricing":{"input":1,"output":5,"cacheRead":0.1,"cacheWrite":1.25,"webSearchPerCall":0.01},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://platform.claude.com/docs/en/about-claude/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["cacheWrite is the 5-minute TTL rate. One-hour cache write: $2 per million tokens. Cache writes replace ordinary input charges, not an additive fee.","Standard global routing; Batch, fast modes and regional processing differ."]},"notes":["Standard Messages output limit; thinking shares output budget."]},{"provider":"xai","id":"grok-4.7","label":"Grok 4.7","selection":"default","contextWindow":500000,"maxOutput":128000,"thinkingMode":"adaptive","reasoningEfforts":["low","medium","high","xhigh"],"skill":7,"description":"xAI flagship for coding and agent work. Same price as 4.6, stronger on tools and harder implementation.","source":{"urls":["https://docs.x.ai/developers/models/grok-4.7","https://docs.x.ai/developers/model-capabilities/images/understanding","https://docs.x.ai/developers/pricing"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"reasoning":true,"imageInput":{"supported":true,"mimeTypes":["image/jpeg","image/png"],"notes":["20 MiB per image; decoded versus encoded accounting unspecified. No image-count limit stated, but context limits still apply.","Hard pixel dimensions and numeric provider resize recommendation unknown."],"sourceUrls":["https://docs.x.ai/developers/models/grok-4.7","https://docs.x.ai/developers/model-capabilities/images/understanding"]}},"clientPolicy":{"maxOutputTokens":128000,"images":{"maxLongEdge":1568,"maxPixels":2500000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":20000000,"maxImages":20,"outputMimeType":"image/png"}},"pricing":{"input":2,"output":6,"cacheRead":0.5,"longContext":{"above":200000,"input":4,"output":12,"cacheRead":1}},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.x.ai/developers/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Long-context rates apply to all tokens in the request when prompt tokens \u003e=200000.","Standard global routing; other service tiers and regions differ. Cache write/read omissions mean Unknown, not free."]},"notes":["Provider documents no separate text output limit. maxOutput=128000 is Pumpkin request policy, not a provider maximum.","Reasoning cannot be disabled."]},{"provider":"xai","id":"grok-4.6","label":"Grok 4.6","selection":"default","contextWindow":500000,"maxOutput":128000,"thinkingMode":"adaptive","reasoningEfforts":["low","medium","high","xhigh"],"skill":6,"description":"Fast workhorse: investigation, evidence collection, intermediate code, image analysis. Good value; less stubborn than Fable on long stuck turns.","source":{"urls":["https://docs.x.ai/developers/models/grok-4.6","https://docs.x.ai/developers/model-capabilities/images/understanding","https://docs.x.ai/developers/pricing"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"reasoning":true,"imageInput":{"supported":true,"mimeTypes":["image/jpeg","image/png"],"notes":["20 MiB per image; decoded versus encoded accounting unspecified. No image-count limit stated, but context limits still apply.","Hard pixel dimensions and numeric provider resize recommendation unknown."],"sourceUrls":["https://docs.x.ai/developers/models/grok-4.6","https://docs.x.ai/developers/model-capabilities/images/understanding"]}},"clientPolicy":{"maxOutputTokens":128000,"images":{"maxLongEdge":1568,"maxPixels":2500000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":20000000,"maxImages":20,"outputMimeType":"image/png"}},"pricing":{"input":2,"output":6,"cacheRead":0.5,"longContext":{"above":200000,"input":4,"output":12,"cacheRead":1}},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.x.ai/developers/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Long-context rates apply to all tokens in the request when prompt tokens \u003e=200000.","Standard global routing; other service tiers and regions differ. Cache write/read omissions mean Unknown, not free."]},"notes":["Provider documents no separate text output limit. maxOutput=128000 is Pumpkin request policy, not a provider maximum.","Reasoning cannot be disabled."]},{"provider":"openai","id":"gpt-5.6-sol","label":"GPT-5.6 Sol","selection":"default","contextWindow":1050000,"maxOutput":128000,"thinkingMode":"adaptive","reasoningEfforts":["none","low","medium","high","xhigh","max"],"skill":5,"description":"Solid mid OpenAI: focused features, bug fixes with a known cause, normal coding loops.","source":{"urls":["https://developers.openai.com/api/docs/models/gpt-5.6-sol","https://developers.openai.com/api/docs/guides/images-vision","https://developers.openai.com/api/docs/pricing"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"streaming":true,"reasoning":true,"imageInput":{"supported":true,"mimeTypes":["image/png","image/jpeg","image/webp","image/gif"],"maxImages":1500,"notes":["GIF must be nonanimated. Total request payload up to 512 MB; byte accounting unspecified.","Original/auto preserves resolution subject to a 65535px preprocessing cap; more than 30000 processed 32px patches per image is rejected. Pumpkin chooses a smaller high-detail patch budget, not original-detail fidelity."],"sourceUrls":["https://developers.openai.com/api/docs/models/gpt-5.6-sol","https://developers.openai.com/api/docs/guides/images-vision"]}},"clientPolicy":{"images":{"maxLongEdge":2048,"maxPixels":2560000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":20000000,"maxImages":20,"outputMimeType":"image/png","patchSize":32,"maxPatches":2500}},"pricing":{"input":4,"output":20,"cacheRead":0.4,"cacheWrite":5,"webSearchPerCall":0.01,"longContext":{"above":272001,"input":8,"output":30,"cacheRead":0.8,"cacheWrite":10}},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://developers.openai.com/api/docs/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Long-context rates apply to all tokens in the request when input tokens \u003e272000 (inclusive lower bound 272001).","Standard global routing; other service tiers and regions differ. Cache write/read omissions mean Unknown, not free."]},"notes":["Maximum input 922000 tokens; reasoning and visible output share the output budget.","Promotional prices at least through 2026-11-21; subsequent rates unknown."]},{"provider":"openai","id":"gpt-5.6-terra","label":"GPT-5.6 Terra","selection":"default","contextWindow":1050000,"maxOutput":128000,"thinkingMode":"adaptive","reasoningEfforts":["none","low","medium","high","xhigh","max"],"skill":4,"description":"Small, well-specified work: one-file edits, straightforward tests, low-stakes helpers.","source":{"urls":["https://developers.openai.com/api/docs/models/gpt-5.6-terra","https://developers.openai.com/api/docs/guides/images-vision","https://developers.openai.com/api/docs/pricing"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"streaming":true,"reasoning":true,"imageInput":{"supported":true,"mimeTypes":["image/png","image/jpeg","image/webp","image/gif"],"maxImages":1500,"notes":["GIF must be nonanimated. Total request payload up to 512 MB; byte accounting unspecified.","Original/auto preserves resolution subject to a 65535px preprocessing cap; more than 30000 processed 32px patches per image is rejected. Pumpkin chooses a smaller high-detail patch budget, not original-detail fidelity."],"sourceUrls":["https://developers.openai.com/api/docs/models/gpt-5.6-terra","https://developers.openai.com/api/docs/guides/images-vision"]}},"clientPolicy":{"images":{"maxLongEdge":2048,"maxPixels":2560000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":20000000,"maxImages":20,"outputMimeType":"image/png","patchSize":32,"maxPatches":2500}},"pricing":{"input":2,"output":12,"cacheRead":0.2,"cacheWrite":2.5,"webSearchPerCall":0.01,"longContext":{"above":272001,"input":4,"output":18,"cacheRead":0.4,"cacheWrite":5}},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://developers.openai.com/api/docs/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Long-context rates apply to all tokens in the request when input tokens \u003e272000 (inclusive lower bound 272001).","Standard global routing; other service tiers and regions differ. Cache write/read omissions mean Unknown, not free."]},"notes":["Maximum input 922000 tokens; reasoning and visible output share the output budget."]},{"provider":"openai","id":"gpt-5.6-luna","label":"GPT-5.6 Luna","selection":"default","contextWindow":1050000,"maxOutput":128000,"thinkingMode":"adaptive","reasoningEfforts":["none","low","medium","high","xhigh","max"],"skill":2,"description":"Cheapest OpenAI: tiny mechanical tasks, classification, short rewrites. Skip anything that needs judgment.","source":{"urls":["https://developers.openai.com/api/docs/models/gpt-5.6-luna","https://developers.openai.com/api/docs/guides/images-vision","https://developers.openai.com/api/docs/pricing"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"streaming":true,"reasoning":true,"imageInput":{"supported":true,"mimeTypes":["image/png","image/jpeg","image/webp","image/gif"],"maxImages":1500,"notes":["GIF must be nonanimated. Total request payload up to 512 MB; byte accounting unspecified.","Original/auto preserves resolution subject to a 65535px preprocessing cap; more than 30000 processed 32px patches per image is rejected. Pumpkin chooses a smaller high-detail patch budget, not original-detail fidelity."],"sourceUrls":["https://developers.openai.com/api/docs/models/gpt-5.6-luna","https://developers.openai.com/api/docs/guides/images-vision"]}},"clientPolicy":{"images":{"maxLongEdge":2048,"maxPixels":2560000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":20000000,"maxImages":20,"outputMimeType":"image/png","patchSize":32,"maxPatches":2500}},"pricing":{"input":0.2,"output":1.2,"cacheRead":0.02,"cacheWrite":0.25,"webSearchPerCall":0.01,"longContext":{"above":272001,"input":0.4,"output":1.8,"cacheRead":0.04,"cacheWrite":0.5}},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://developers.openai.com/api/docs/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Long-context rates apply to all tokens in the request when input tokens \u003e272000 (inclusive lower bound 272001).","Standard global routing; other service tiers and regions differ. Cache write/read omissions mean Unknown, not free."]},"notes":["Maximum input 922000 tokens; reasoning and visible output share the output budget."]},{"provider":"openai","id":"gpt-6-astra","label":"GPT-6 Astra","selection":"default","contextWindow":1050000,"maxOutput":128000,"thinkingMode":"adaptive","reasoningEfforts":["low","medium","high","xhigh","max"],"skill":10,"description":"OpenAI flagship: hard implementation and long context, same class as Fable. Use when the problem is large or ambiguous.","source":{"urls":["https://developers.openai.com/api/docs/models/gpt-6-astra","https://developers.openai.com/api/docs/guides/images-vision","https://developers.openai.com/api/docs/pricing"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"streaming":true,"reasoning":true,"imageInput":{"supported":true,"mimeTypes":["image/png","image/jpeg","image/webp","image/gif"],"maxImages":1500,"notes":["GIF must be nonanimated. Total request payload up to 512 MB; byte accounting unspecified.","Original/auto preserves resolution subject to a 65535px preprocessing cap; more than 30000 processed 32px patches per image is rejected. Pumpkin chooses a smaller high-detail patch budget, not original-detail fidelity."],"sourceUrls":["https://developers.openai.com/api/docs/models/gpt-6-astra","https://developers.openai.com/api/docs/guides/images-vision"]}},"clientPolicy":{"images":{"maxLongEdge":65535,"maxPixels":2560000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":20000000,"maxImages":20,"outputMimeType":"image/png","patchSize":32,"maxPatches":2500}},"pricing":{"input":10,"output":50,"cacheRead":1,"cacheWrite":12.5,"webSearchPerCall":0.01,"longContext":{"above":272001,"input":20,"output":75,"cacheRead":2,"cacheWrite":25}},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://developers.openai.com/api/docs/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Long-context rates apply to all tokens in the request when input tokens \u003e272000 (inclusive lower bound 272001).","Standard global routing; other service tiers and regions differ. Cache write/read omissions mean Unknown, not free."]},"notes":["Maximum input 922000 tokens; reasoning and visible output share the output budget.","Tool calling requires Responses; Chat Completions supports text only."]},{"provider":"meta","id":"muse-spark-1.3","label":"Muse Spark 1.3","selection":"default","contextWindow":1048576,"maxOutput":65536,"thinkingMode":"adaptive","reasoningEfforts":["minimal","low","medium","high","xhigh","max"],"skill":6,"description":"Meta's agentic reasoning model: multi-step tool loops and coding, 1M context, cheap cached input.","source":{"urls":["https://dev.meta.ai/docs/models","https://dev.meta.ai/docs/pricing-rate-limits","https://dev.meta.ai/docs/reasoning"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"streaming":true,"reasoning":true,"imageInput":{"supported":true,"mimeTypes":["image/png","image/jpeg","image/webp","image/gif"],"maxImages":20,"notes":["Images as base64 data URLs or Files API references; audio understanding is degraded on 1.3 (use 1.2), and GIF uses the first frame.","Meta documents no per-image or per-request image byte limits; Pumpkin applies its own conservative client policy rather than trusting an unspecified vendor cap."],"sourceUrls":["https://dev.meta.ai/docs/models","https://dev.meta.ai/docs/image-understanding"]}},"clientPolicy":{"images":{"maxLongEdge":1568,"maxPixels":2500000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":20000000,"maxImages":20,"outputMimeType":"image/png","patchSize":32,"maxPatches":2500}},"pricing":{"input":1.25,"output":4.25,"cacheRead":0.15,"webSearchPerCall":0.0025},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://dev.meta.ai/docs/pricing-rate-limits"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard tier (prompts not used for training). The -contributor tier is cheaper but trains on prompts and is deliberately not cataloged.","No long-context premium: the same rate applies at any context fill. Cached input is $0.15/M; Meta lists no separate cache-write price (omission means Unknown, not free)."]},"notes":["Reasoning cannot be turned off (effort \"none\" returns HTTP 400); reasoning shares the output budget with visible text.","Meta documents no per-request output ceiling; 65536 is a safe request cap below the 1,048,576 context window (input and output share one budget)."]},{"provider":"fireworks","id":"accounts/fireworks/models/deepseek-v4-pro-0813","label":"DeepSeek V4 Pro 0813","selection":"opt-in","contextWindow":1048576,"maxOutput":16384,"thinkingMode":"none","skill":6,"description":"Strongest DeepSeek: hard coding, agents and long reasoning; Fireworks' top open pick.","source":{"urls":["https://app.fireworks.ai/models/fireworks/deepseek-v4-pro-0813","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":false,"notes":["Discovery reports image input unsupported; curated support is conservatively disabled."],"sourceUrls":["https://api.fireworks.ai/v1/accounts/fireworks/models"]}},"clientPolicy":{"maxOutputTokens":16384},"pricing":{"input":1.32,"output":3.96,"cacheRead":0.044},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.fireworks.ai/serverless/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"deprecatedAt":"2026-09-25","notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits.","Serverless retirement announced for 2026-09-25; exact cutoff time unknown. Dedicated deployments are excluded."]},{"provider":"fireworks","id":"accounts/fireworks/models/kimi-k3","label":"Kimi K3","selection":"opt-in","contextWindow":1048576,"maxOutput":16384,"thinkingMode":"none","skill":6,"description":"Kimi K3 flagship: hard coding and research agents, image input, 1M context; the priciest here.","source":{"urls":["https://fireworks.ai/models/fireworks/kimi-k3","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":true,"mimeTypes":["image/png","image/jpeg","image/gif","image/bmp","image/tiff","image/x-portable-pixmap"],"maxImages":30,"notes":["Aggregate base64 image data must be less than 10 MB; MB definition unspecified. URL images less than 5 MB and download within 1.5 seconds.","No numeric universal provider resize recommendation. Pumpkin uses its conservative resize policy."],"sourceUrls":["https://fireworks.ai/models/fireworks/kimi-k3","https://docs.fireworks.ai/guides/querying-vision-language-models"]}},"clientPolicy":{"maxOutputTokens":16384,"images":{"maxLongEdge":1568,"maxPixels":2500000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":9000000,"maxImages":20,"outputMimeType":"image/png"}},"pricing":{"input":3,"output":15,"cacheRead":0.3},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.fireworks.ai/serverless/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning."]},{"provider":"fireworks","id":"accounts/fireworks/models/glm-5p3","label":"GLM 5.3","selection":"opt-in","contextWindow":1048576,"maxOutput":16384,"thinkingMode":"none","skill":6,"description":"Newest GLM flagship: coding, reasoning and agent tool use; 1M context.","source":{"urls":["https://fireworks.ai/models/fireworks/glm-5p3","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":false,"notes":["Discovery reports image input unsupported; curated support is conservatively disabled."],"sourceUrls":["https://api.fireworks.ai/v1/accounts/fireworks/models"]}},"clientPolicy":{"maxOutputTokens":16384},"pricing":{"input":1.4,"output":4.4,"cacheRead":0.26},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.fireworks.ai/serverless/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits."]},{"provider":"fireworks","id":"accounts/fireworks/models/glm-5p2","label":"GLM 5.2","selection":"opt-in","contextWindow":1048576,"maxOutput":16384,"thinkingMode":"none","skill":5,"description":"GLM 5.2 flagship: coding and agent tool use; 1M context.","source":{"urls":["https://fireworks.ai/models/fireworks/glm-5p2","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":false,"notes":["Discovery reports image input unsupported; curated support is conservatively disabled."],"sourceUrls":["https://api.fireworks.ai/v1/accounts/fireworks/models"]}},"clientPolicy":{"maxOutputTokens":16384},"pricing":{"input":1.4,"output":4.4,"cacheRead":0.14},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.fireworks.ai/serverless/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"deprecatedAt":"2026-09-25","notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits.","Serverless retirement announced for 2026-09-25; exact cutoff time unknown. Dedicated deployments are excluded."]},{"provider":"fireworks","id":"accounts/fireworks/models/kimi-k2p7-code","label":"Kimi K2.7 Code","selection":"opt-in","contextWindow":262144,"maxOutput":16384,"thinkingMode":"none","skill":6,"description":"Kimi K2.7 Code: coding-tuned agent model with image input; 262K context.","source":{"urls":["https://fireworks.ai/models/fireworks/kimi-k2p7-code","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":true,"mimeTypes":["image/png","image/jpeg","image/gif","image/bmp","image/tiff","image/x-portable-pixmap"],"maxImages":30,"notes":["Aggregate base64 image data must be less than 10 MB; MB definition unspecified. URL images less than 5 MB and download within 1.5 seconds.","No numeric universal provider resize recommendation. Pumpkin uses its conservative resize policy."],"sourceUrls":["https://fireworks.ai/models/fireworks/kimi-k2p7-code","https://docs.fireworks.ai/guides/querying-vision-language-models"]}},"clientPolicy":{"maxOutputTokens":16384,"images":{"maxLongEdge":1568,"maxPixels":2500000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":9000000,"maxImages":20,"outputMimeType":"image/png"}},"pricing":{"input":0.95,"output":4,"cacheRead":0.19},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.fireworks.ai/serverless/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"deprecatedAt":"2026-09-25","notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits.","Serverless retirement announced for 2026-09-25; exact cutoff time unknown. Dedicated deployments are excluded."]},{"provider":"fireworks","id":"accounts/fireworks/models/kimi-k2p6","label":"Kimi K2.6","selection":"opt-in","contextWindow":262144,"maxOutput":16384,"thinkingMode":"none","skill":5,"description":"Kimi K2.6: agents with tool use, images and documents; 262K context.","source":{"urls":["https://fireworks.ai/models/fireworks/kimi-k2p6","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":true,"mimeTypes":["image/png","image/jpeg","image/gif","image/bmp","image/tiff","image/x-portable-pixmap"],"maxImages":30,"notes":["Aggregate base64 image data must be less than 10 MB; MB definition unspecified. URL images less than 5 MB and download within 1.5 seconds.","No numeric universal provider resize recommendation. Pumpkin uses its conservative resize policy."],"sourceUrls":["https://fireworks.ai/models/fireworks/kimi-k2p6","https://docs.fireworks.ai/guides/querying-vision-language-models"]}},"clientPolicy":{"maxOutputTokens":16384,"images":{"maxLongEdge":1568,"maxPixels":2500000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":9000000,"maxImages":20,"outputMimeType":"image/png"}},"pricing":{"input":0.95,"output":4,"cacheRead":0.16},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.fireworks.ai/serverless/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"deprecatedAt":"2026-09-25","notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits.","Serverless retirement announced for 2026-09-25; exact cutoff time unknown. Dedicated deployments are excluded."]},{"provider":"fireworks","id":"accounts/fireworks/models/qwen3p8-2p4t-a95b","label":"Qwen 3.8 2.4T A95B","selection":"opt-in","contextWindow":262144,"maxOutput":16384,"thinkingMode":"none","skill":5,"description":"Qwen 3.8 (2.4T MoE, 95B active): strong general coding and reasoning; 262K context.","source":{"urls":["https://fireworks.ai/models/fireworks/qwen3p8-2p4t-a95b","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":false,"notes":["Discovery reports image input unsupported; curated support is conservatively disabled."],"sourceUrls":["https://api.fireworks.ai/v1/accounts/fireworks/models"]}},"clientPolicy":{"maxOutputTokens":16384},"availability":{"status":"unavailable","fetchedAt":"2026-09-23T10:44:43Z"},"notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits.","Public model page says serverless not supported. No substitution of qwen3p8-max pricing."]},{"provider":"fireworks","id":"accounts/fireworks/models/minimax-m3","label":"MiniMax M3","selection":"opt-in","contextWindow":512000,"maxOutput":16384,"thinkingMode":"none","skill":5,"description":"MiniMax M3: fast coding and agent model, good value; 512K context.","source":{"urls":["https://fireworks.ai/models/fireworks/minimax-m3","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":false,"notes":["Discovery reports image input unsupported; curated support is conservatively disabled."],"sourceUrls":["https://api.fireworks.ai/v1/accounts/fireworks/models"]}},"clientPolicy":{"maxOutputTokens":16384},"pricing":{"input":0.3,"output":1.2,"cacheRead":0.06},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.fireworks.ai/serverless/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits.","Exact serving-model feature metadata says no images despite generic multimodal architecture prose."]},{"provider":"fireworks","id":"accounts/fireworks/models/deepseek-v4p1-flash","label":"DeepSeek V4.1 Flash","selection":"opt-in","contextWindow":1048576,"maxOutput":16384,"thinkingMode":"none","skill":4,"description":"DeepSeek V4.1 Flash with image input: quick coding and agent steps at low cost; 1M context.","source":{"urls":["https://app.fireworks.ai/models/fireworks/deepseek-v4p1-flash","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":true,"mimeTypes":["image/png","image/jpeg","image/gif","image/bmp","image/tiff","image/x-portable-pixmap"],"maxImages":30,"notes":["Aggregate base64 image data must be less than 10 MB; MB definition unspecified. URL images less than 5 MB and download within 1.5 seconds.","No numeric universal provider resize recommendation. Pumpkin uses its conservative resize policy."],"sourceUrls":["https://app.fireworks.ai/models/fireworks/deepseek-v4p1-flash","https://docs.fireworks.ai/guides/querying-vision-language-models"]}},"clientPolicy":{"maxOutputTokens":16384,"images":{"maxLongEdge":1568,"maxPixels":2500000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":9000000,"maxImages":20,"outputMimeType":"image/png"}},"pricing":{"input":0.22,"output":0.66,"cacheRead":0.007},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.fireworks.ai/serverless/pricing","https://app.fireworks.ai/models/fireworks/deepseek-v4p1-flash","https://docs.fireworks.ai/updates/changelog"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"pricingSchedule":[{"effectiveFrom":"2026-10-01T00:00:00Z","pricing":{"input":0.3,"output":1.2,"cacheRead":0.006},"revision":"2026-10-01"}],"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits."]},{"provider":"fireworks","id":"accounts/fireworks/models/nemotron-3-ultra-nvfp4","label":"Nemotron 3 Ultra NVFP4","selection":"opt-in","contextWindow":262144,"maxOutput":16384,"thinkingMode":"none","skill":4,"description":"NVIDIA Nemotron 3 Ultra (preview): large reasoning model; 262K context.","source":{"urls":["https://fireworks.ai/models/fireworks/nemotron-3-ultra-nvfp4","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":false,"notes":["Discovery reports image input unsupported; curated support is conservatively disabled."],"sourceUrls":["https://api.fireworks.ai/v1/accounts/fireworks/models"]}},"clientPolicy":{"maxOutputTokens":16384},"pricing":{"input":0.6,"output":2.4,"cacheRead":0.12},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.fireworks.ai/serverless/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits."]},{"provider":"fireworks","id":"accounts/fireworks/models/deepseek-v4-flash-0731","label":"DeepSeek V4 Flash 0731","selection":"opt-in","contextWindow":1048576,"maxOutput":16384,"thinkingMode":"none","skill":3,"description":"Fast, cheap DeepSeek: extraction, summaries and everyday agent steps; 1M context.","source":{"urls":["https://app.fireworks.ai/models/fireworks/deepseek-v4-flash-0731","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":false,"notes":["Discovery reports image input unsupported; curated support is conservatively disabled."],"sourceUrls":["https://api.fireworks.ai/v1/accounts/fireworks/models"]}},"clientPolicy":{"maxOutputTokens":16384},"pricing":{"input":0.22,"output":0.66,"cacheRead":0.007},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.fireworks.ai/serverless/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"deprecatedAt":"2026-09-25","notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits.","Serverless retirement announced for 2026-09-25; exact cutoff time unknown. Dedicated deployments are excluded."]},{"provider":"fireworks","id":"accounts/fireworks/models/deepseek-v4-flash-vision-exp","label":"DeepSeek V4 Flash Vision Experimental","selection":"opt-in","contextWindow":1048576,"maxOutput":16384,"thinkingMode":"none","skill":3,"description":"DeepSeek V4 Flash with experimental image input; cheap, 1M context.","source":{"urls":["https://app.fireworks.ai/models/fireworks/deepseek-v4-flash-vision-exp","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":true,"mimeTypes":["image/png","image/jpeg","image/gif","image/bmp","image/tiff","image/x-portable-pixmap"],"maxImages":30,"notes":["Aggregate base64 image data must be less than 10 MB; MB definition unspecified. URL images less than 5 MB and download within 1.5 seconds.","No numeric universal provider resize recommendation. Pumpkin uses its conservative resize policy."],"sourceUrls":["https://app.fireworks.ai/models/fireworks/deepseek-v4-flash-vision-exp","https://docs.fireworks.ai/guides/querying-vision-language-models"]}},"clientPolicy":{"maxOutputTokens":16384,"images":{"maxLongEdge":1568,"maxPixels":2500000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":9000000,"maxImages":20,"outputMimeType":"image/png"}},"pricing":{"input":0.22,"output":0.66,"cacheRead":0.007},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.fireworks.ai/serverless/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"deprecatedAt":"2026-09-25","notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits.","Serverless retirement announced for 2026-09-25; exact cutoff time unknown. Dedicated deployments are excluded."]},{"provider":"fireworks","id":"accounts/fireworks/models/gpt-oss-120b","label":"GPT OSS 120B","selection":"opt-in","contextWindow":131072,"maxOutput":16384,"thinkingMode":"none","skill":3,"description":"OpenAI open-weight 120B: solid reasoning and tool use at low cost; no images.","source":{"urls":["https://fireworks.ai/models/fireworks/gpt-oss-120b","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":false,"notes":["Discovery reports image input unsupported; curated support is conservatively disabled."],"sourceUrls":["https://api.fireworks.ai/v1/accounts/fireworks/models"]}},"clientPolicy":{"maxOutputTokens":16384},"pricing":{"input":0.15,"output":0.6,"cacheRead":0.015},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.fireworks.ai/serverless/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits."]},{"provider":"fireworks","id":"accounts/fireworks/models/glm-5p3-flash","label":"GLM 5.3 Flash","selection":"opt-in","contextWindow":1048576,"maxOutput":16384,"thinkingMode":"none","skill":3,"description":"Small, cheap GLM with image input: quick edits and lookups.","source":{"urls":["https://fireworks.ai/models/fireworks/glm-5p3-flash","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":true,"mimeTypes":["image/png","image/jpeg","image/gif","image/bmp","image/tiff","image/x-portable-pixmap"],"maxImages":30,"notes":["Aggregate base64 image data must be less than 10 MB; MB definition unspecified. URL images less than 5 MB and download within 1.5 seconds.","No numeric universal provider resize recommendation. Pumpkin uses its conservative resize policy."],"sourceUrls":["https://fireworks.ai/models/fireworks/glm-5p3-flash","https://docs.fireworks.ai/guides/querying-vision-language-models"]}},"clientPolicy":{"maxOutputTokens":16384,"images":{"maxLongEdge":1568,"maxPixels":2500000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":9000000,"maxImages":20,"outputMimeType":"image/png"}},"pricing":{"input":0.15,"output":0.5,"cacheRead":0.03},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.fireworks.ai/serverless/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits."]},{"provider":"fireworks","id":"accounts/fireworks/models/muse-glimmer-30b","label":"Muse Glimmer 30B","selection":"opt-in","contextWindow":131072,"maxOutput":16384,"thinkingMode":"none","skill":3,"description":"Small 30B always-on agent model with image input: cheap and quick for simple tasks.","source":{"urls":["https://fireworks.ai/models/fireworks/muse-glimmer-30b","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":true,"mimeTypes":["image/png","image/jpeg","image/gif","image/bmp","image/tiff","image/x-portable-pixmap"],"maxImages":30,"notes":["Aggregate base64 image data must be less than 10 MB; MB definition unspecified. URL images less than 5 MB and download within 1.5 seconds.","No numeric universal provider resize recommendation. Pumpkin uses its conservative resize policy."],"sourceUrls":["https://fireworks.ai/models/fireworks/muse-glimmer-30b","https://docs.fireworks.ai/guides/querying-vision-language-models"]}},"clientPolicy":{"maxOutputTokens":16384,"images":{"maxLongEdge":1568,"maxPixels":2500000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":9000000,"maxImages":20,"outputMimeType":"image/png"}},"pricing":{"input":0.35,"output":1.5,"cacheRead":0.04},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.fireworks.ai/serverless/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"deprecatedAt":"2026-09-25","notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits.","Serverless retirement announced for 2026-09-25; exact cutoff time unknown. Dedicated deployments are excluded."]},{"provider":"fireworks","id":"accounts/fireworks/models/nemotron-lightning-3p5-30b-a3b","label":"Nemotron Lightning 3.5 30B A3B","selection":"opt-in","contextWindow":262144,"maxOutput":16384,"thinkingMode":"none","skill":2,"description":"Cheapest fast agent model (30B, 3B active): small edits, descriptions, triage.","source":{"urls":["https://fireworks.ai/models/fireworks/nemotron-lightning-3p5-30b-a3b","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":false,"notes":["Discovery reports image input unsupported; curated support is conservatively disabled."],"sourceUrls":["https://api.fireworks.ai/v1/accounts/fireworks/models"]}},"clientPolicy":{"maxOutputTokens":16384},"pricing":{"input":0.05,"output":0.2,"cacheRead":0.01},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://docs.fireworks.ai/serverless/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits."]},{"provider":"fireworks","id":"accounts/fireworks/models/inkling","label":"Inkling","selection":"opt-in","contextWindow":1048576,"maxOutput":16384,"thinkingMode":"none","description":"Fireworks serverless model with image input and a 1M context; capability not rated.","source":{"urls":["https://fireworks.ai/models/fireworks/inkling","https://docs.fireworks.ai/guides/querying-vision-language-models","https://docs.fireworks.ai/serverless/pricing","https://docs.fireworks.ai/updates/changelog","https://api.fireworks.ai/v1/accounts/fireworks/models"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"imageInput":{"supported":true,"mimeTypes":["image/png","image/jpeg","image/gif","image/bmp","image/tiff","image/x-portable-pixmap"],"maxImages":30,"notes":["Aggregate base64 image data must be less than 10 MB; MB definition unspecified. URL images less than 5 MB and download within 1.5 seconds.","No numeric universal provider resize recommendation. Pumpkin uses its conservative resize policy."],"sourceUrls":["https://fireworks.ai/models/fireworks/inkling","https://docs.fireworks.ai/guides/querying-vision-language-models"]}},"clientPolicy":{"maxOutputTokens":16384,"images":{"maxLongEdge":1568,"maxPixels":2500000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":9000000,"maxImages":20,"outputMimeType":"image/png"}},"pricing":{"input":1,"output":4.05,"cacheRead":0.17},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://fireworks.ai/models/fireworks/inkling"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Standard global serverless rates. Priority, Fast and regional prices differ. No separately verified cache-write rate."]},"availability":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"},"notes":["Provider output limit Unknown. thinkingMode=none and the 16384 output cap describe Pumpkin adapter policy, not absence of provider reasoning.","Exact context Unknown until metadata discovery; rounded public context labels are not treated as exact limits."]},{"provider":"openai","id":"gpt-6-sol","label":"GPT-6 Sol","selection":"default","contextWindow":1050000,"maxOutput":128000,"thinkingMode":"adaptive","reasoningEfforts":["none","low","medium","high","xhigh","max"],"skill":6,"description":"Complex coding and agent work when Astra is more than you need. Cheaper than GPT-5.6 Sol.","source":{"urls":["https://developers.openai.com/api/docs/models/gpt-6-sol","https://developers.openai.com/api/docs/guides/images-vision","https://developers.openai.com/api/docs/pricing"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"streaming":true,"reasoning":true,"imageInput":{"supported":true,"mimeTypes":["image/png","image/jpeg","image/webp","image/gif"],"maxImages":1500,"notes":["GIF must be nonanimated. The image guide allows up to 512 MB total request payload; exact byte accounting is unspecified. Model-specific image dimensions remain Unknown."],"sourceUrls":["https://developers.openai.com/api/docs/models/gpt-6-sol","https://developers.openai.com/api/docs/guides/images-vision"]}},"clientPolicy":{"images":{"maxLongEdge":1568,"maxPixels":2500000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":20000000,"maxImages":20,"outputMimeType":"image/png"}},"pricing":{"input":2,"output":10,"cacheRead":0.2,"cacheWrite":2.5,"webSearchPerCall":0.01,"longContext":{"above":272001,"input":4,"output":15,"cacheRead":0.4,"cacheWrite":5}},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://developers.openai.com/api/docs/models/gpt-6-sol","https://developers.openai.com/api/docs/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Long-context rates apply to all tokens in the request when input tokens \u003e272000 (inclusive lower bound 272001).","Cache writes are 1.25x uncached input; cache reads are 0.1x. Batch/Flex are half Standard; Fast is twice the applicable rates. Regional processing adds 10% where available. EU data residency requires Standard processing."]},"notes":["Use Responses for built-in tools and function calling. Chat Completions function calling requires reasoning_effort=none.","Exact model-specific image resize limits are not stated in the reviewed image guide; Pumpkin uses its conservative client policy, not an inferred provider maximum."]},{"provider":"openai","id":"gpt-6-luna","label":"GPT-6 Luna","selection":"default","contextWindow":1050000,"maxOutput":128000,"thinkingMode":"adaptive","reasoningEfforts":["none","low","medium","high","xhigh","max"],"skill":3,"description":"Cheapest OpenAI: high-volume tasks, classification, short rewrites. Not for design or long hunts.","source":{"urls":["https://developers.openai.com/api/docs/models/gpt-6-luna","https://developers.openai.com/api/docs/guides/images-vision","https://developers.openai.com/api/docs/pricing"],"verifiedAt":"2026-09-22T00:00:00Z"},"capabilities":{"tools":true,"streaming":true,"reasoning":true,"imageInput":{"supported":true,"mimeTypes":["image/png","image/jpeg","image/webp","image/gif"],"maxImages":1500,"notes":["GIF must be nonanimated. The image guide allows up to 512 MB total request payload; exact byte accounting is unspecified. Model-specific image dimensions remain Unknown."],"sourceUrls":["https://developers.openai.com/api/docs/models/gpt-6-luna","https://developers.openai.com/api/docs/guides/images-vision"]}},"clientPolicy":{"images":{"maxLongEdge":1568,"maxPixels":2500000,"maxDecodedBytes":100000000,"maxEncodedBytesTotal":20000000,"maxImages":20,"outputMimeType":"image/png"}},"pricing":{"input":0.1,"output":0.5,"cacheRead":0.01,"cacheWrite":0.125,"webSearchPerCall":0.01,"longContext":{"above":272001,"input":0.2,"output":0.75,"cacheRead":0.02,"cacheWrite":0.25}},"pricingMetadata":{"currency":"USD","unitTokens":1000000,"serviceTier":"standard","sourceUrls":["https://developers.openai.com/api/docs/models/gpt-6-luna","https://developers.openai.com/api/docs/pricing"],"verifiedAt":"2026-09-22T00:00:00Z","revision":"2026-09-22","notes":["Long-context rates apply to all tokens in the request when input tokens \u003e272000 (inclusive lower bound 272001).","Cache writes are 1.25x uncached input; cache reads are 0.1x. Batch/Flex are half Standard; Fast is twice the applicable rates. Regional processing adds 10% where available. EU data residency requires Standard processing."]},"notes":["Use Responses for built-in tools and function calling. Chat Completions function calling requires reasoning_effort=none.","Exact model-specific image resize limits are not stated in the reviewed image guide; Pumpkin uses its conservative client policy, not an inferred provider maximum."]}],"providerStatus":{"fireworks":{"status":"available","fetchedAt":"2026-09-23T10:44:43Z"}}}