DevPass (@LLMDevPass) header
DevPass (@LLMDevPass) profile picture

DevPass

@LLMDevPass

One AI coding subscription. Every model. by @llmgateway

 Earth
devpass.llmgateway.io
Joined August 2026
  • Tweets283
  • Following30
  • Followers123
73Media
Photo by DevPass (@LLMDevPass)Photo by DevPass (@LLMDevPass)Photo by DevPass (@LLMDevPass)Photo by DevPass (@LLMDevPass)Photo by DevPass (@LLMDevPass)Photo by DevPass (@LLMDevPass)Photo by DevPass (@LLMDevPass)Photo by DevPass (@LLMDevPass)Photo by DevPass (@LLMDevPass)Photo by DevPass (@LLMDevPass)
One in ten requests. Nearly half the tokens. DeepSeek V4.1 Flash 1017 took 10.0% of LLM Gateway's calls in 24h and 47.8% of its 44.8B tokens: 177.2k per call vs 37.2k board-wide. The original V4.1 Flash had more calls (205.5k) at 46.7k each.
Stacked bar chart of hourly token volume on LLM Gateway over the last 24 hours: 44.8B tokens, 1.2M requests, 150 models ranked; DeepSeek V4.1 Flash 1017 leads with 47.8% of tokens, followed by DeepSeek V4.1 Flash at 21.4% and Claude Opus 5.5 at 4.5%.

ALT Stacked bar chart of hourly token volume on LLM Gateway over the last 24 hours: 44.8B tokens, 1.2M requests, 150 models ranked; DeepSeek V4.1 Flash 1017 leads with 47.8% of tokens, followed by DeepSeek V4.1 Flash at 21.4% and Claude Opus 5.5 at 4.5%.

1
1
2
77
Same family, 3.8x the tokens per call. Biggest 500M+ riser: GPT-6.1 Sol, +35.8% to 1.9B. All 150 models, ranked live: llmgateway.io/rankings
LLM Rankings — Top Models by Real Usage

LLM Rankings — Top Models by Real Usage

Live LLM rankings from real traffic routed through LLM Gateway: top models and apps by token volume, usage trends, and provider market share.

llmgateway.io
4
Unpopular opinion: annual billing on AI coding tools is lock-in sold as a discount. Models turn over in weeks. Your prepaid year doesn't. DevPass: monthly only, cancel anytime. $29 → $58 of usage at provider rates, up to 8% of it per rolling 24h. devpass.llmgateway.io
DevPass by LLM Gateway - All-Access Dev Plans for AI Coding

DevPass by LLM Gateway - All-Access Dev Plans for AI Coding

One subscription, every coding model. Fixed-price dev plans for Claude Code, Cursor, and 100+ models.

devpass.llmgateway.io
11
GLM-5.3 FlashX is live on LLM Gateway: Z.ai's multimodal GLM with reasoning and tool calling. 1M context, vision + tools. $0.37/M in, $1.25/M out. llmgateway.io/models/glm-5.3…
34
Step 5 Preview is live on LLM Gateway: StepFun's multimodal reasoning preview. 1M context, vision + tools + reasoning, structured output. $1/M in, $2.70/M out. llmgateway.io/models/step-5-…
Step 5 Preview — Pricing, Providers & Benchmarks

Step 5 Preview — Pricing, Providers & Benchmarks

StepFun's multimodal reasoning preview model with a 1M-token context window, structured output, and function calling. Use Step 5 Preview via one OpenAI-compatible API with live per-token pricing...

llmgateway.io
1
30
Two DeepSeek V4.1 Flash builds took 68.1% of LLM Gateway's tokens in 24h, and both grew: the 1017 build +58.0% to 21.6B, the original +42.0% to 12.2B. Fastest 500M+ riser: Claude Opus 5.5, +62.6% to 2.0B. GLM-5.3 fell 79.0%, from 2.4B to 499.7M. 49.6B tokens, 1.3M requests.
LLM Gateway rankings for the last 24 hours: 49.6B tokens, 1.3M requests and 155 models, with an hourly stacked bar chart of the top 8 models led by DeepSeek V4.1 Flash 1017 (43.5%), DeepSeek V4.1 Flash (24.6%) and Claude Opus 5.5 (4.1%).

ALT LLM Gateway rankings for the last 24 hours: 49.6B tokens, 1.3M requests and 155 models, with an hourly stacked bar chart of the top 8 models led by DeepSeek V4.1 Flash 1017 (43.5%), DeepSeek V4.1 Flash (24.6%) and Claude Opus 5.5 (4.1%).

1
83
GLM-5.3 slipped off the chart to #10. All 155 models, re-ranked every few minutes: llmgateway.io/rankings
LLM Rankings — Top Models by Real Usage

LLM Rankings — Top Models by Real Usage

Live LLM rankings from real traffic routed through LLM Gateway: top models and apps by token volume, usage trends, and provider market share.

llmgateway.io
12
Kimi K2.8 Preview is live on LLM Gateway: Moonshot's preview model with vision, reasoning and tool calling. 1M context. $1/M in, $4/M out. llmgateway.io/models/kimi-k2…
Kimi K2.8 Preview — Pricing, Providers & Benchmarks

Kimi K2.8 Preview — Pricing, Providers & Benchmarks

Moonshot's Kimi K2.8 preview model with vision, reasoning, tool calling, and a 1M-token context window. Use Kimi K2.8 Preview via one OpenAI-compatible API with live per-token pricing across provid...

llmgateway.io
1
2
71
FLUX 3 Max is live on LLM Gateway: Black Forest Labs' video model, served via fal. Text-to-video and image-to-video, 5-20s clips at 720p or 1080p, optional audio. $0.17/sec at 720p, $0.29/sec at 1080p. llmgateway.io/models/flux-3-…
FLUX 3 Max — Pricing, Providers & Benchmarks

FLUX 3 Max — Pricing, Providers & Benchmarks

FLUX 3 Max text-to-video and image-to-video generation with optional audio. Use FLUX 3 Max via one OpenAI-compatible API with live per-token pricing across providers.

llmgateway.io
24
DeepSeek V4.1 Flash Cyber is live on LLM Gateway for enterprise plans: built for security tooling and agents. 1M context, vision + tools + reasoning, $0.30/M in, $1/M out. llmgateway.io/models/deepsee…
DeepSeek V4.1 Flash Cyber — Pricing, Providers & Benchmarks

DeepSeek V4.1 Flash Cyber — Pricing, Providers & Benchmarks

Abliterated DeepSeek V4.1 Flash 1017 for security tooling and agents, with fewer refusals, vision, tool calling, and a 1M-token context window. Enterprise plan only.

llmgateway.io
1
4
593
The full GLM-5.3 overtook its own Flash sibling on LLM Gateway in 24h. GLM-5.3: +112.8% to 2.4B tokens GLM-5.3 Flash: -43.2% to 2.3B Biggest 500M+ mover: MiniMax M3.1 Flash, +690.5% to 1.2B. 43.0B tokens, 1.4M requests. DeepSeek V4.1 Flash 1017 still #1 at 32.2%.
LLM Gateway rankings for the last 24 hours: 43B tokens processed, 1.4M requests routed and 130 models ranked, with an hourly stacked bar chart of token volume led by DeepSeek V4.1 Flash 1017 (32.2%), DeepSeek V4.1 Flash (20.4%) and GPT-6 Luna (6.1%).

ALT LLM Gateway rankings for the last 24 hours: 43B tokens processed, 1.4M requests routed and 130 models ranked, with an hourly stacked bar chart of token volume led by DeepSeek V4.1 Flash 1017 (32.2%), DeepSeek V4.1 Flash (20.4%) and GPT-6 Luna (6.1%).

1
1
1
95
Full 24h board, all 130 models, refreshed every few minutes: llmgateway.io/rankings
LLM Rankings — Top Models by Real Usage

LLM Rankings — Top Models by Real Usage

Live LLM rankings from real traffic routed through LLM Gateway: top models and apps by token volume, usage trends, and provider market share.

llmgateway.io
12
Cloudflare Workers AI is live on LLM Gateway: DeepSeek V4.1 Flash 1017, V4 Pro, V4 Flash, GLM-5.3 and GLM-5.3 Flash, all with 1M context. GLM-5.3 Flash from $0.15/M in, $0.50/M out. llmgateway.io/providers/clou…
Cloudflare API — Models & Pricing | LLM Gateway

Cloudflare API — Models & Pricing | LLM Gateway

Access 5 Cloudflare models through LLM Gateway's OpenAI-compatible API with per-token pricing, automatic fallback, caching, and cost analytics.

llmgateway.io
1
1
101
$29 → $58 $79 → $158 $179 → $358 DevPass: plan price → monthly usage at provider list rates, no token markup. Published limits: 8-10% of the allowance per rolling 24h, 10-15% a week on premium models. Max includes 2 Reset Passes a month. devpass.llmgateway.io
DevPass by LLM Gateway - All-Access Dev Plans for AI Coding

DevPass by LLM Gateway - All-Access Dev Plans for AI Coding

One subscription, every coding model. Fixed-price dev plans for Claude Code, Cursor, and 100+ models.

devpass.llmgateway.io
23
Three of LLM Gateway's top 6 models roughly tripled their tokens in 24h: Claude Opus 5.5 +215.9% to 1.2B GLM-5.3 Flash +200.0% to 4.1B, now at #3 GPT-6.1 Sol +196.7% to 1.4B The two DeepSeek V4.1 Flash builds barely moved (+3.6%, +3.2%) and still took 55.7% of 43.6B.
LLM Gateway rankings for the last 24 hours: 43.6B tokens, 1.2M requests, 129 models ranked, shown as hourly stacked bars led by DeepSeek V4.1 Flash 1017 (31.5%), DeepSeek V4.1 Flash (24.2%) and GLM-5.3 Flash (9.5%).

ALT LLM Gateway rankings for the last 24 hours: 43.6B tokens, 1.2M requests, 129 models ranked, shown as hourly stacked bars led by DeepSeek V4.1 Flash 1017 (31.5%), DeepSeek V4.1 Flash (24.2%) and GLM-5.3 Flash (9.5%).

1
2
4
141
Any model shortlist has a shelf life. The 121 models below the chart, ranked live: llmgateway.io/rankings
LLM Rankings — Top Models by Real Usage

LLM Rankings — Top Models by Real Usage

Live LLM rankings from real traffic routed through LLM Gateway: top models and apps by token volume, usage trends, and provider market share.

llmgateway.io
10
VEED Fabric 1.0 and Fabric 1.0 Fast are live on LLM Gateway: talking-video generation. One face image plus one speech track becomes a lip-synced video, as long as the audio. 480p or 720p, from $0.088/sec. llmgateway.io/models/veed-fa…
VEED Fabric 1.0 — Pricing, Providers & Benchmarks

VEED Fabric 1.0 — Pricing, Providers & Benchmarks

VEED Fabric 1.0 talking-video generation. Animates one image with a face into a lip-synced video driven by one speech audio track; the output is as long as the audio.

llmgateway.io
1
1
35
Gemini Omni 1.1 Flash is live on LLM Gateway: Google's video model with native audio. 360p to 4K, 3-10s clips, plus edit and extend on existing video. From $0.034/sec at 360p, $0.10/sec at 720p. llmgateway.io/models/gemini-…
Gemini Omni 1.1 Flash — Pricing, Providers & Benchmarks

Gemini Omni 1.1 Flash — Pricing, Providers & Benchmarks

Google Gemini Omni 1.1 Flash video generation with native audio. Use Gemini Omni 1.1 Flash via one OpenAI-compatible API with live per-token pricing across providers.

llmgateway.io
2
2
110
KLING v3.0 Pro is live on LLM Gateway: 1080p text-to-video and image-to-video, 5s or 10s clips, with or without audio. $0.112/sec silent, $0.168/sec with audio. llmgateway.io/models/kling-v…
KLING v3.0 Pro — Pricing, Providers & Benchmarks

KLING v3.0 Pro — Pricing, Providers & Benchmarks

AtlasCloud KLING v3.0 Pro video generation. Routes text and image requests to the matching upstream KLING v3.0 Pro model at 1080p. Use KLING v3.0 Pro via one OpenAI-compatible API with live per-token...

llmgateway.io
2
2
75
Qwen3.8 Omni Flash is live on LLM Gateway: Alibaba's omni-modal model that understands text, images and audio. 1M context, tools + adjustable reasoning. $0.15/M in, $0.47/M out. llmgateway.io/models/qwen3.8…
Qwen3.8 Omni Flash — Pricing, Providers & Benchmarks

Qwen3.8 Omni Flash — Pricing, Providers & Benchmarks

Omni-modal Qwen3.8 Flash model that understands text, images and audio, with adjustable reasoning, tool use and a 1M context. Use Qwen3.8 Omni Flash via one OpenAI-compatible API with live per-token...

llmgateway.io
1
1
41
Load more