Rewrite BitRouter vs OpenRouter, fix stale platform-fee claims - #62
Merged
Merged
Conversation
The page argued on agent-native primitives (KYA, x402, injection checks) and a ~5ms vs ~30ms latency gap — neither is a real differentiator against OpenRouter, which is also low-latency and agent-native. Rewrote it around the three things that actually differ: we're fully open source and self-hostable, we take no cut, and the router learns your workload. The intro now credits OpenRouter for building the category before pivoting, rather than opening with an enumeration the bolded lead-ins already cover. Fee claims were contradictory across the repo: the pricing page, models page and llms-shared.ts all said 0% markup, while three CompareTable rows and the migration guide said "2% stablecoin / 5% card". 0% is correct; fixed all four stale spots. The vs-OpenRouter prose now concedes that OpenRouter passes inference through at 0% too, and names the two fees that genuinely differ (5.5% to load credits, 5% on BYOK). Cut the comparison table from 10 rows to 5 — one per argument. Ten rows was longer than the prose it supported, repeated the "when OpenRouter wins" section, and invited row-counting against a catalog we don't try to match. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
SPIKESPIGEL404
added a commit
that referenced
this pull request
Aug 7, 2026
PR #62 landed the OpenRouter rewrite and its own platform-fee fix, hitting the same registry file. Resolution: - LiteLLM rows: ours (the 5-row trim is what this PR is for) - OpenRouter rows: main's — #62 rewrote that comparison deliberately, and it is not this PR's business - Portkey fee row: main's "none · 0% markup" phrasing, now the house style, superseding the equivalent change made here Net effect: this branch's diff against main is now only the LiteLLM entry. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Rewrites
/docs/overview/bitrouter-vs-openrouteras a short, conversational page, and fixes a pricing claim that was contradicting itself across the repo.The page
The old version argued on agent-native primitives (KYA, x402, MCP gateway, injection checks) and a ~5ms vs ~30ms latency gap. Neither holds up as a differentiator — OpenRouter is also low-latency and agent-native — so both are gone. The page now argues the three things that actually differ:
And three honest reasons to pick OpenRouter instead: catalog breadth, free-tier models, and enterprise depth.
The intro opens by crediting OpenRouter for building the category before pivoting to what we're betting on — the enumeration it replaced was scaffolding the bolded lead-ins already provided.
The pricing fix
The repo said two different things about what BitRouter charges:
lib/llms-shared.ts, enterprise and startup pagesCompareTablerows and the migration guide0% is correct. Fixed all four stale spots (the OpenRouter, LiteLLM, and Portkey comparison rows, plus
migrate-from-openrouter.mdx). Left alone, the vs-OpenRouter page would have claimed 0% in prose with a table two inches below saying 2%.The prose also concedes that OpenRouter passes inference through at 0% too, then names the fees that genuinely differ — 5.5% to load credits, 5% on BYOK past their free allowance. Conceding the shared ground is what makes the rest credible to a reader who already knows their pricing.
The table
Cut from 10 rows to 5, one per argument. Ten rows ran longer than the prose it supported, restated the "when OpenRouter wins" section, and turned the page into a scorecard against a catalog we deliberately don't try to match.
Verified
pnpm lint:docspasses (49 docs)tscshows only pre-existing errors incomponents/ai/search.tsxandcomponents/markdown.tsxBefore merging
🤖 Generated with Claude Code