{"slug": "openrouter-fusion", "name": "OpenRouter Fusion", "category": "ai", "category_name": "AI tools and apps", "description": "", "recommenders": 3, "commenters": 10, "warners": 2, "score": 1.912, "rising": 0.35, "first": "2026-06-15", "latest": "2026-08-14", "why": "I play around with OpenRouter Fusion some, and it does a pretty good job", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 3, "neutral": 0, "compare": 5, "warn": 2}, "spark": [0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0], "listed": true, "loved": [], "caveat": "", "short_quote": "I play around with OpenRouter Fusion some, and it does a pretty good job", "kind": "", "icon": true, "domain": "openrouter.ai", "github_repo": null, "official_url": "https://openrouter.ai", "pricing": "", "aliases": ["OpenRouter Fusion"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 788733040, "item_id": 49295815, "url": "https://news.ycombinator.com/item?id=49295815", "by": "mirekrusin", "date": "2026-08-14", "version": null, "stance": "recommend", "quote": "ie. open router fusion", "excerpt": "ie. open router fusion [0] [0] https://openrouter.ai/openrouter/fusion", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 3.7 Flash"}, {"id": 784872016, "item_id": 49054501, "url": "https://news.ycombinator.com/item?id=49054501", "by": "cortical_iv", "date": "2026-07-25", "version": null, "stance": "recommend", "quote": "I play around with OpenRouter Fusion some, and it does a pretty good job", "excerpt": "This is a very cool nontrivial problem. I worked on this some in a mixture of agent type context. There were lots of failure modes mostly involving model aggregators not working well because they tend to end up plagiarizing what they consider the \"best\" LLM in the target pool (that is, they don't actually give a balanced view of the target results, but more a winner take all). I had to build all sorts of tooling to a", "context": "", "use_case": "Multi-model answer aggregation", "switched_from": "", "story": "Collective Intelligence: The Next Frontier of AI"}, {"id": 776632720, "item_id": 48539545, "url": "https://news.ycombinator.com/item?id=48539545", "by": "eknkc", "date": "2026-06-15", "version": null, "stance": "recommend", "quote": "But this feels like it would work great on some situations.", "excerpt": "I opened the page and prompted it `Which 3d printer is the best`. I mean this is a stupid question but I was looking at some 3d printers so it popped into my mind. Seeing this log is interesting: https://link.ekin.dev/6RzYGGX7 It came up with a decent response but I guess Opus or GPT 5.5 would do fine anyway. Gotta try it on different stuff. But this feels like it would work great on some situations.", "context": "", "use_case": "Multi-model API", "switched_from": "", "story": "Openrouter Fusion API"}]}], "caveats": [{"label": "", "summary": "", "quotes": [{"id": 776635376, "item_id": 48539711, "url": "https://news.ycombinator.com/item?id=48539711", "by": "arizen", "date": "2026-06-15", "version": null, "stance": "warn", "quote": "I run same query I used for Fable on OR Fusion and results were worse.", "excerpt": "Some anecdata on Fusion: I run same query I used for Fable on OR Fusion and results were worse. It felt, like Fable was able to kinda grasp very deep knowledge/intelligence layers and outline solution not only in agreeable way, but rather it proposed to prioritize solution items, with discarding some of the items, which made a lot of sense to me. While Fusion felt more like a bit diversified answer of the same class ", "context": "", "use_case": "multi-model queries", "switched_from": "", "story": "Openrouter Fusion API"}, {"id": 776630832, "item_id": 48539427, "url": "https://news.ycombinator.com/item?id=48539427", "by": "michaelbuckbee", "date": "2026-06-15", "version": null, "stance": "warn", "quote": "Fusion was 7x slower and 4x the cost.", "excerpt": "I ran a quick eval to see what this looks like qualitatively vs just calling Opus 4.7 or GPT 5.5 directly. As expected, Fusion was 7x slower and 4x the cost. https://3fpi5avcqq.evvl.io/", "context": "", "use_case": "", "switched_from": "", "story": "Openrouter Fusion API"}]}], "who": [], "tags": [], "use_cases": ["multi-model answer aggregation", "multi-model queries", "multi-model api"], "compared": [{"slug": "gpt-5", "name": "GPT-5", "co": 2, "switched_from": 0, "switched_to": 0}, {"slug": "claude-opus", "name": "Claude Opus", "co": 2, "switched_from": 0, "switched_to": 0}], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=49026810", "title": "Show HN: Echo \u2013 Fable-level results at 1/3 the cost using open-weight models", "mentions": 3}, {"url": "https://news.ycombinator.com/item?id=48537641", "title": "Openrouter Fusion API", "mentions": 3}, {"url": "https://news.ycombinator.com/item?id=49099143", "title": "Launch HN: Tokenless (YC S26) \u2013 Automatic model switching to save money", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=49289112", "title": "Gemini 3.7 Flash", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49053465", "title": "Collective Intelligence: The Next Frontier of AI", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48624782", "title": "Sakana Fugu", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {}, "2026-01": {}, "2026-02": {}, "2026-03": {}, "2026-04": {}, "2026-05": {}, "2026-06": {"warn": 2, "recommend": 1, "compare": 1}, "2026-07": {"compare": 4, "recommend": 1}, "2026-08": {"recommend": 1}, "2026-09": {}, "2026-10": {}}, "versions": [], "version_unnamed": 0, "family_ref": null, "runs_on": null, "lines": [], "quotes": [{"id": 788733040, "item_id": 49295815, "url": "https://news.ycombinator.com/item?id=49295815", "by": "mirekrusin", "date": "2026-08-14", "version": null, "stance": "recommend", "quote": "ie. open router fusion", "excerpt": "ie. open router fusion [0] [0] https://openrouter.ai/openrouter/fusion", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 3.7 Flash"}, {"id": 785601872, "item_id": 49100117, "url": "https://news.ycombinator.com/item?id=49100117", "by": "Barbing", "date": "2026-07-29", "version": null, "stance": "compare", "quote": "", "excerpt": "Can you tell if this site is doing two and three? https://pellmell.ai via https://news.ycombinator.com/item?id=48980458 (incl. example chat) and just now found the developer\u2019s explanation https://news.ycombinator.com/item?id=49028740: aimed at improving performance without added latency but from a different angle. Instead of waiting for all replies for synthesis (like OpenRouter Fusion), it streams the \"best\" reply i", "context": "", "use_case": "", "switched_from": "", "story": "Launch HN: Tokenless (YC S26) \u2013 Automatic model switching to save money"}, {"id": 785596832, "item_id": 49099802, "url": "https://news.ycombinator.com/item?id=49099802", "by": "random17", "date": "2026-07-29", "version": null, "stance": "compare", "quote": "are different from ours in that", "excerpt": "These approaches, along with OpenRouter Fusion (https://openrouter.ai/blog/announcements/fusion-beats-frontier/) are different from ours in that: 1. we trained custom models to predict LLM performance (this performs much better than asking LLMs directly to predict performance) 2. we cut off model responses early if they are not performing well, saving on costs versus running them to completion 3. finally we serve the", "context": "", "use_case": "", "switched_from": "", "story": "Launch HN: Tokenless (YC S26) \u2013 Automatic model switching to save money"}, {"id": 784872016, "item_id": 49054501, "url": "https://news.ycombinator.com/item?id=49054501", "by": "cortical_iv", "date": "2026-07-25", "version": null, "stance": "recommend", "quote": "I play around with OpenRouter Fusion some, and it does a pretty good job", "excerpt": "This is a very cool nontrivial problem. I worked on this some in a mixture of agent type context. There were lots of failure modes mostly involving model aggregators not working well because they tend to end up plagiarizing what they consider the \"best\" LLM in the target pool (that is, they don't actually give a balanced view of the target results, but more a winner take all). I had to build all sorts of tooling to a", "context": "", "use_case": "Multi-model answer aggregation", "switched_from": "", "story": "Collective Intelligence: The Next Frontier of AI"}, {"id": 784459840, "item_id": 49028740, "url": "https://news.ycombinator.com/item?id=49028740", "by": "seizethecheese", "date": "2026-07-23", "version": null, "stance": "compare", "quote": "Fusion is a totally different approach", "excerpt": "Fusion is a totally different approach, though similar in the sense that it leverages different models. Fusion generates many replies then synthesizes. This adds a ton of latency and cost, so it's going to be better only for cases where you're willing to wait a lot and pay a lot more. Routers (like this project) are a different thing, they can theoretically improve performance and cost at the same time without increa", "context": "", "use_case": "", "switched_from": "", "story": "Show HN: Echo \u2013 Fable-level results at 1/3 the cost using open-weight models"}, {"id": 784452096, "item_id": 49028256, "url": "https://news.ycombinator.com/item?id=49028256", "by": "meander_water", "date": "2026-07-23", "version": null, "stance": "compare", "quote": "Seems similar to Openrouter Fusion", "excerpt": "Seems similar to Openrouter Fusion - https://openrouter.ai/docs/guides/routing/routers/fusion-router", "context": "", "use_case": "", "switched_from": "", "story": "Show HN: Echo \u2013 Fable-level results at 1/3 the cost using open-weight models"}, {"id": 784441969, "item_id": 49027623, "url": "https://news.ycombinator.com/item?id=49027623", "by": "seizethecheese", "date": "2026-07-23", "version": null, "stance": "compare", "quote": "instead of waiting for all models to synthesize, we stream the best model immediately", "excerpt": "You might be interested in a project that I'm working on, which is kind of like OpenRouter Fusion, but instead of waiting for all models to synthesize, we stream the best model immediately and background the rest. The background models then reconcile with an emoji reaction and optional reply. It gets similar results to Fusion and is a lot faster! There's a free version that leverages open weight models here: http://p", "context": "", "use_case": "", "switched_from": "", "story": "Show HN: Echo \u2013 Fable-level results at 1/3 the cost using open-weight models"}, {"id": 778004368, "item_id": 48625273, "url": "https://news.ycombinator.com/item?id=48625273", "by": "stygiansonic", "date": "2026-06-21", "version": null, "stance": "compare", "quote": "Looks like Fusion calls a bunch of models and then uses an LLM to synthesize the results", "excerpt": "From a brief reading of what Fusion does: https://openrouter.ai/docs/guides/features/plugins/fusion Looks like Fusion calls a bunch of models and then uses an LLM to synthesize the results, and pass to another model for final output. Fugu looks like it's doing something different? Using an LLM earlier on in the flow as an orchestrator to decide which other LLMs to call. More coordinator than simply synthesizing resul", "context": "", "use_case": "", "switched_from": "", "story": "Sakana Fugu"}, {"id": 776635376, "item_id": 48539711, "url": "https://news.ycombinator.com/item?id=48539711", "by": "arizen", "date": "2026-06-15", "version": null, "stance": "warn", "quote": "I run same query I used for Fable on OR Fusion and results were worse.", "excerpt": "Some anecdata on Fusion: I run same query I used for Fable on OR Fusion and results were worse. It felt, like Fable was able to kinda grasp very deep knowledge/intelligence layers and outline solution not only in agreeable way, but rather it proposed to prioritize solution items, with discarding some of the items, which made a lot of sense to me. While Fusion felt more like a bit diversified answer of the same class ", "context": "", "use_case": "multi-model queries", "switched_from": "", "story": "Openrouter Fusion API"}, {"id": 776632720, "item_id": 48539545, "url": "https://news.ycombinator.com/item?id=48539545", "by": "eknkc", "date": "2026-06-15", "version": null, "stance": "recommend", "quote": "But this feels like it would work great on some situations.", "excerpt": "I opened the page and prompted it `Which 3d printer is the best`. I mean this is a stupid question but I was looking at some 3d printers so it popped into my mind. Seeing this log is interesting: https://link.ekin.dev/6RzYGGX7 It came up with a decent response but I guess Opus or GPT 5.5 would do fine anyway. Gotta try it on different stuff. But this feels like it would work great on some situations.", "context": "", "use_case": "Multi-model API", "switched_from": "", "story": "Openrouter Fusion API"}, {"id": 776630832, "item_id": 48539427, "url": "https://news.ycombinator.com/item?id=48539427", "by": "michaelbuckbee", "date": "2026-06-15", "version": null, "stance": "warn", "quote": "Fusion was 7x slower and 4x the cost.", "excerpt": "I ran a quick eval to see what this looks like qualitatively vs just calling Opus 4.7 or GPT 5.5 directly. As expected, Fusion was 7x slower and 4x the cost. https://3fpi5avcqq.evvl.io/", "context": "", "use_case": "", "switched_from": "", "story": "Openrouter Fusion API"}]}