RouteMux (@RouteMux) header
RouteMux (@RouteMux) profile picture

RouteMux

@RouteMux

One API. Every top model, only success is billed.

 Monrovia, CA
routemux.com
Joined September 2010
  • Tweets290
  • Following115
  • Followers54
27Media
Photo by RouteMux (@RouteMux)Photo by RouteMux (@RouteMux)Photo by RouteMux (@RouteMux)Photo by RouteMux (@RouteMux)Photo by RouteMux (@RouteMux)Photo by RouteMux (@RouteMux)Photo by RouteMux (@RouteMux)Photo by RouteMux (@RouteMux)Photo by RouteMux (@RouteMux)Photo by RouteMux (@RouteMux)
New models and price cuts, up to 85% off official prices. Cheaper: Grok 4.6 / 4.7 at 85% off ($0.30 / $0.90 per 1M), DeepSeek V4.1 Flash at half price. New: Gemini 4 Argon, Qwen3.8, MiniMax M3.1 Flash, Muse Spark 1.3, Nano Banana 2.1 / Pro, Grok Imagine image + video.
RouteMux new models and price cuts, October 2026. Grok 4.6 / 4.7 85% off, Gemini 4 Argon 50% off, Qwen3.8 30% off, Nano Banana Pro 25% off, Grok Imagine up to 70% off, MiniMax H3 video 50% off, MiniMax M3.1 Flash 70% off, DeepSeek V4.1 Flash 50% off. Up to 85% off official prices: Grok 4.6 / 4.7 at $0.30 / $0.90 per 1M tokens in / out (official $2 / $6), DeepSeek V4.1 Flash at $0.15 / $0.60 (official $0.30 / $1.20).

ALT RouteMux new models and price cuts, October 2026. Grok 4.6 / 4.7 85% off, Gemini 4 Argon 50% off, Qwen3.8 30% off, Nano Banana Pro 25% off, Grok Imagine up to 70% off, MiniMax H3 video 50% off, MiniMax M3.1 Flash 70% off, DeepSeek V4.1 Flash 50% off. Up to 85% off official prices: Grok 4.6 / 4.7 at $0.30 / $0.90 per 1M tokens in / out (official $2 / $6), DeepSeek V4.1 Flash at $0.15 / $0.60 (official $0.30 / $1.20).

2
1
113
MiniMax H3-Image and H3-Image Highspeed are live on RouteMux at 50% off official token pricing: - Standard: ~$0.01 per 1K image, ~$0.03 per 4K image ($1.15 text in / $0.75 image in / $1.90 image out per 1M tokens) - Highspeed: ~$0.065 per 4K image ($1.15 text in / $1.50 image in / $3.80 image out per 1M tokens) Engineering details: - Native 4096x4096 (16 MP) and custom sizes up to 20 MP - Latency: Highspeed generates a 4K frame in ~8s (~14s end-to-end via RouteMux including full-res relay; ~20s on Standard) - Editing: up to 16 HTTP(S) or data-URL reference images and up to 10 outputs per OpenAI Images API call Try in Playground or check API specs: routemux.com/minimax/minimax…
MiniMax H3-Image on RouteMux: 4K image generation in about 8 seconds on Highspeed, 50% off official token pricing.

ALT MiniMax H3-Image on RouteMux: 4K image generation in about 8 seconds on Highspeed, 50% off official token pricing.

1
1
67
Claude Fable 5.2 is live on RouteMux at 65% off Anthropic list pricing: $2.80 input / $14.00 output per 1M tokens (official $8 / $40), with cache reads at $0.07. Specs & migration notes from Fable 5.1: - 1M context window, 128K max output, always-on adaptive thinking (low to max effort) - Preserved-thinking prefix checks now apply to every account: keep history append-only or set prefix_mismatch_behavior to "drop_block" - Legacy computer_20251124 returns 400; move to computer_toolset_20260801 - RouteMux strips unsupported sampling params (temperature, top_p, top_k) automatically on /v1/messages Specs & live pricing: routemux.com/anthropic/claud…
Claude Fable 5.2 on RouteMux: 65% off Anthropic list price, $2.80 input / $14 output / $0.07 cache read per 1M tokens, 1M context, 128K max output.

ALT Claude Fable 5.2 on RouteMux: 65% off Anthropic list price, $2.80 input / $14 output / $0.07 cache read per 1M tokens, 1M context, 128K max output.

38
Full tables, protocols and top-up notes: routemux.com/announcements/a…
A wave of new models, and price cuts | RouteMux

A wave of new models, and price cuts | RouteMux

RouteMux has recently added a batch of new text, image and video models, and several models now cost less. Price cuts come first, then each new model's pr…

routemux.com
10
Old names still work, no code changes, routed to the newer model at its price: DeepSeek V4 Flash / Vision Exp → V4.1 Flash: $0.15 / $0.60, was $0.44 / $1.32 Grok 4.6 / 4.7: now 85% off official, $0.30 / $0.90 (was $0.40 / $1.20) Grok 4.5 → 4.7
Price cuts. DeepSeek V4.1 Flash is half price: $0.30 / $1.20 to $0.15 / $0.60 per 1M tokens. DeepSeek V4 Flash and V4 Flash Vision Exp now route to V4.1 Flash: $0.44 / $1.32 to $0.15 / $0.60. MiniMax H3 video: 768p $0.08 to $0.04 per second, 2K $0.13 to $0.065. Grok 4.7 and 4.6: $0.40 / $1.20 to $0.30 / $0.90.

ALT Price cuts. DeepSeek V4.1 Flash is half price: $0.30 / $1.20 to $0.15 / $0.60 per 1M tokens. DeepSeek V4 Flash and V4 Flash Vision Exp now route to V4.1 Flash: $0.44 / $1.32 to $0.15 / $0.60. MiniMax H3 video: 768p $0.08 to $0.04 per second, 2K $0.13 to $0.065. Grok 4.7 and 4.6: $0.40 / $1.20 to $0.30 / $0.90.

1
17
Video per second, all 50% off official: Grok Imagine Video 1.5: 480p $0.04 · 720p $0.07 · 1080p $0.125, with first/last frames, keyframes and references. MiniMax H3 is now four models: H3 from $0.04, Highspeed $0.03, Lite from $0.0125, Max from $0.025, all with references.
Video prices per second, against official prices, all 50% off. Grok Imagine Video 1.5: 480p $0.04 vs $0.08, 720p $0.07 vs $0.14, 1080p $0.125 vs $0.25; each input image $0.005 vs $0.01. MiniMax H3: 768p $0.04 vs $0.08, 2K $0.065 vs $0.13. MiniMax H3 Highspeed, new: 768p $0.03 vs $0.06. MiniMax H3 Lite, new: 480p $0.0125 vs $0.025, 768p $0.02 vs $0.04. MiniMax H3 Max, new: 480p $0.025 vs $0.05, 768p $0.04 vs $0.08.

ALT Video prices per second, against official prices, all 50% off. Grok Imagine Video 1.5: 480p $0.04 vs $0.08, 720p $0.07 vs $0.14, 1080p $0.125 vs $0.25; each input image $0.005 vs $0.01. MiniMax H3: 768p $0.04 vs $0.08, 2K $0.065 vs $0.13. MiniMax H3 Highspeed, new: 768p $0.03 vs $0.06. MiniMax H3 Lite, new: 480p $0.0125 vs $0.025, 768p $0.02 vs $0.04. MiniMax H3 Max, new: 480p $0.025 vs $0.05, 768p $0.04 vs $0.08.

1
17
Image models, per image: Nano Banana 2.1: 1K $0.025 · 2K $0.04 · 4K $0.08 Nano Banana Pro: 1K/2K $0.10 · 4K $0.18, 25% off Grok Imagine Image: $0.006, 70% off Grok Imagine Image 2.0: $0.012–$0.024, 70% off, edits GPT Image 2.5 Sunburst & Flare: now 2K $0.08 · 4K $0.12
Image model prices per image against official prices. Nano Banana 2.1: 1K $0.025 vs $0.0336, 2K $0.04 vs $0.0504, 4K $0.08 vs $0.113. Nano Banana Pro: 1K and 2K $0.10 vs $0.134, 4K $0.18 vs $0.24, 25% off. Grok Imagine Image $0.006 vs $0.02, 70% off. Grok Imagine Image 2.0 $0.012 to $0.024 vs $0.04 to $0.08, 70% off. GPT Image 2.5 Sunburst and Flare at 2K $0.08 and 4K $0.12, no official per-image price.

ALT Image model prices per image against official prices. Nano Banana 2.1: 1K $0.025 vs $0.0336, 2K $0.04 vs $0.0504, 4K $0.08 vs $0.113. Nano Banana Pro: 1K and 2K $0.10 vs $0.134, 4K $0.18 vs $0.24, 25% off. Grok Imagine Image $0.006 vs $0.02, 70% off. Grok Imagine Image 2.0 $0.012 to $0.024 vs $0.04 to $0.08, 70% off. GPT Image 2.5 Sunburst and Flare at 2K $0.08 and 4K $0.12, no official per-image price.

1
27
Before you switch to Muse Spark 1.3 Contributor: it is cheap because Meta may train on your requests and responses. Don't send it anything sensitive. It also doesn't speak the Anthropic Messages protocol yet, so Claude Code can't use it. OpenAI Chat and Responses both work.
1
32
Text, USD per 1M tokens in / out: Gemini 4 Argon: $1.00 / $5.00, 50% off Qwen3.8-Max: $1.40 / $4.20, 30% off Qwen3.8-Flash: $0.105 / $0.329, 30% off MiniMax M3.1 Flash: $0.045 / $0.18, 70% off Muse Spark 1.3: $0.05 / $0.10, 50% off DeepSeek V4.1 Flash: $0.15 / $0.60, 50% off
Text model prices per 1M tokens, input / output, against official prices. Gemini 4 Argon $1.00 / $5.00 vs $2.00 / $10.00, 50% off. Qwen3.8-Max $1.40 / $4.20 vs $2.00 / $6.00, 30% off. Qwen3.8-Flash $0.105 / $0.329 vs $0.15 / $0.47, 30% off. MiniMax M3.1 Flash $0.045 / $0.18 vs $0.15 / $0.60, 70% off. Muse Spark 1.3 Contributor $0.05 / $0.10 vs $0.10 / $0.20, 50% off. DeepSeek V4.1 Flash $0.15 / $0.60 vs $0.30 / $1.20, 50% off.

ALT Text model prices per 1M tokens, input / output, against official prices. Gemini 4 Argon $1.00 / $5.00 vs $2.00 / $10.00, 50% off. Qwen3.8-Max $1.40 / $4.20 vs $2.00 / $6.00, 30% off. Qwen3.8-Flash $0.105 / $0.329 vs $0.15 / $0.47, 30% off. MiniMax M3.1 Flash $0.045 / $0.18 vs $0.15 / $0.60, 70% off. Muse Spark 1.3 Contributor $0.05 / $0.10 vs $0.10 / $0.20, 50% off. DeepSeek V4.1 Flash $0.15 / $0.60 vs $0.30 / $1.20, 50% off.

1
1
59
GPT-6.1 Sol is 85% off official pricing on RouteMux: $0.30 / $1.50 per 1M input/output tokens (official: $2 / $10). One API key, 40+ models, 10 providers: Claude, GPT, Gemini, Grok, Qwen, Kimi, DeepSeek and more. OpenAI & Anthropic compatible. Prices: routemux.com/models?utm_sour…
1
45
Until Oct 6 00:00 UTC: all GPT models at half our regular price, and +20% credit on every top-up. Per 1M in/out: GPT-6.1 Sol $0.20/$1.00 → $0.10/$0.50 Claude Sonnet 5.5 $0.56/$2.80 → $0.40/$2.00 (after a top-up) Not affiliated with OpenAI or Anthropic. routemux.com/pricing?utm_sou…
RouteMux October Sale: +20% credit on every top-up, any amount, never expires. Sale prices per 1M tokens, input / output: GPT-6.1 Sol $0.10 / $0.50; Claude Sonnet 5.5 $0.40 / $2.00. Oct 1 to Oct 6, 00:00 UTC. RouteMux is an independent API gateway.

ALT RouteMux October Sale: +20% credit on every top-up, any amount, never expires. Sale prices per 1M tokens, input / output: GPT-6.1 Sol $0.10 / $0.50; Claude Sonnet 5.5 $0.40 / $2.00. Oct 1 to Oct 6, 00:00 UTC. RouteMux is an independent API gateway.

1
35,201
The +20% applies to every top-up, any amount, with no upper limit. It follows when you place the order and doesn't stack with the first top-up bonus — you get the higher one. Claude and Luna unlock once your account has topped up.
14
Everything in the sale, off list price: · GPT-5.5, GPT-5.6 Sol / Terra, GPT-6 Astra / Sol, GPT-6.1 Sol: 90% → 95% off · GPT-6 Luna, GPT-5.6 Luna: 70% → 85% off · All 11 Claude models: 72% → 80% off
67
🔥 October Sale is live: +20% credit on every top-up, any amount, and it never expires. 📉 Plus five days of lower prices: GPT: 90% → 95% off OpenAI list (GPT-6.1 Sol $0.10 / $0.50 per 1M) Claude: 72% → 80% off list ⏰ Ends Oct 6, 00:00 UTC routemux.com/pricing?utm_sou…
RouteMux October Sale: every GPT model goes from 90% to 95% off OpenAI list price, Claude from 72% to 80% off, +20% credit on every top-up. Oct 1 to Oct 6, 00:00 UTC.

ALT RouteMux October Sale: every GPT model goes from 90% to 95% off OpenAI list price, Claude from 72% to 80% off, +20% credit on every top-up. Oct 1 to Oct 6, 00:00 UTC.

2
144
October Sale starts tomorrow at 00:00 UTC, for five days: half-price GPT (90% → 95% off OpenAI list), Claude from 72% to 80% off, and +20% credit on every top-up. Ends Oct 6, 00:00 UTC. routemux.com/pricing?utm_sou…
Image from RouteMux (@RouteMux)
73
Before you switch: - All three take Anthropic Messages, so switching is a model-name change. GPT-6.1 Sol also takes OpenAI Chat/Responses. - GPT-6.1 Sol has no reasoning "none"; effort is low to max. - GPT-6.1 Sol needs no top-up. The Claude models unlock after your first one.
81
New on RouteMux this past week: GPT-6.1 Sol at 90% off list price, Claude Sonnet 5.5 and Opus 5.5 at 72% off. Per 1M tokens, in / out (list): GPT-6.1 Sol $0.20 / $1.00 ($2 / $10) Sonnet 5.5 $0.56 / $2.80 ($2 / $10) Opus 5.5 $1.12 / $5.60 ($4 / $20) routemux.com/models?utm_sour…
RouteMux pricing card: 3 new models, up to 90% off list price, all on the Anthropic Messages API. Prices per 1M tokens, RouteMux vs official. GPT-6.1 Sol: input $0.20 vs $2, output $1.00 vs $10, cache read $0.01 vs $0.10, 90% off. Claude Sonnet 5.5: input $0.56 vs $2, output $2.80 vs $10, cache read $0.056 vs $0.20, 72% off. Claude Opus 5.5: input $1.12 vs $4, output $5.60 vs $20, cache read $0.056 vs $0.20, 72% off. Model drops from Sep 23 to Sep 30, 2026.

ALT RouteMux pricing card: 3 new models, up to 90% off list price, all on the Anthropic Messages API. Prices per 1M tokens, RouteMux vs official. GPT-6.1 Sol: input $0.20 vs $2, output $1.00 vs $10, cache read $0.01 vs $0.10, 90% off. Claude Sonnet 5.5: input $0.56 vs $2, output $2.80 vs $10, cache read $0.056 vs $0.20, 72% off. Claude Opus 5.5: input $1.12 vs $4, output $5.60 vs $20, cache read $0.056 vs $0.20, 72% off. Model drops from Sep 23 to Sep 30, 2026.

1
102
Input side, per 1M tokens: Opus 5.5: $1.12 (official $4, Opus 5 was $5) GPT-6 Sol: $0.20 (official $2, 5.6 Sol was $4). No top-up needed. GPT-6 Luna: $0.03 (official $0.10, 5.6 Luna was $0.20) routemux.com/anthropic/claud…

Claude Opus 5.5 API — first model of the Claude 5.5 family | RouteMux

Claude Opus 5.5 (Sep 22, 2026) succeeds Opus 5 and is Anthropic's suggested starting point. Thinking is always on, effort defaults to medium, forced tools fail.

routemux.com
49
Anthropic cut Opus by 20%. OpenAI cut GPT-6 by 50%+. We cut again. Output $/1M tokens: old official → new → RouteMux Claude Opus 5.5: $25 → $20 → $5.60 GPT-6 Sol: $20 → $10 → $1.00 GPT-6 Luna: $1.20 → $0.50 → $0.15 Up to 95% below last-gen pricing. routemux.com/openai/gpt-6-so…
Image from RouteMux (@RouteMux)
1
65
On reliability, the measurement rather than the adjective: 143,556 GPT requests over the last 30 days, 98.95% completed. Most of the 1,513 that didn't were interrupted streams, not upstream refusals. Per-model availability is public: routemux.com/status

RouteMux Status — Live model and platform availability

Availability for every model RouteMux serves, measured by our own probes on a fixed schedule. Not inferred from user traffic.

routemux.com
8
Load more