PRICING · MATCH-OR-BETTER · THUMBS-DOWN REFUND

Change two env vars. Cut the bill.

We never charge more than the provider would charge you direct. Usually less — Atlas routes to a cheaper variant whenever quality holds, and you keep the savings. Don’t like a call? Thumbs-down it and we credit it back automatically, no questions asked. Start on free models with no card at all.

export OPENAI_BASE_URL=https://api.newmen.ai/v1 (or ANTHROPIC_BASE_URL=https://api.newmen.ai) plus a Newmen API key. That’s the migration. Card statement descriptor: NEWMEN.AI*CREDITS.

Atlas-1

The smart-routing default.

Atlas-1 is a router, not a single model. It looks at the operation you tagged, the eval-gate history for that operation, and the provider variants available for the underlying model — then picks the cheapest path that has historically stayed green.

- model: "chatgpt-5.5",
+ model: "atlas-1",            // smart routing across providers
+ tier:  "standard",           // optional; default is "standard"
  metadata: { operation_id: "summarize_ticket" },

What you keep

OpenAI-compatible API, your existing SDK, every supported provider model. Pin a specific model anytime.

What changes

atlas-1 picks the upstream + provider variant + quantization per call. The response carries a `delivery` block telling you exactly what was used.

What protects quality

If `metadata.operation_id` resolves to an operation with a bound evaluator and a min_score, calls scoring below threshold aren't billed.

Three tiers

Default Standard. Tighten or loosen as needed.

Tier is a per-call hint. Atlas-1 honours it; pinned models honour it within the variants the upstream provides.

realtime

Realtime

p95 < 300ms TTFT

Cost + 10%

Direct passthrough to the upstream provider, full precision. The only tier with a hard delivery guarantee on closed-weight models. Use when latency is non-negotiable.

standard

Standard

p95 < 8s TTFT

30–55% off

Atlas routes to the cheapest provider variant whose quantization has stayed green for your operation's eval gates. Default tier — your bill drops without changing business logic. Best-effort: may upgrade to Realtime when no path passes; bill follows actual delivery.

batch

Batch

~24h SLA

50–70% off

Async. Routes to provider batch APIs (OpenAI / Anthropic) where supported, queued spot capacity for open-weight models. Largest discounts. Streaming not supported.

Discounts are computed against the realtime cost-leader for the same logical model. Per-task numbers are published openly on the benchmarks page; per-call savings vary with prompt mix and time of day.

Worked example

Summarization workload, $X → $0.4X.

A customer running 30M tokens / month of support-ticket summarization on chatgpt-5.5 (input $2.75 / output $11 per 1M) pays roughly $210 / month.

They flip model: "atlas-1" and add a regex evaluator with min_score: 0.95 to the summarize_ticket operation. After two weeks of green eval gates on Standard tier, Atlas routes ~80% of calls to meta-llama/llama-3.1-70b-instruct Q8 on Hyperbolic ($0.40 / $0.40 per 1M) and ~20% to ChatGPT-5.5 on calls the loop says still need it.

New monthly cost: ~$84. That’s a 60% reduction. Calls scoring below the regex threshold aren’t metered, so the customer absorbs zero quality risk.

Numbers are illustrative. Run the comparison runner on your own prompts for a real per-workload estimate.

The guarantee

If quality drops, the call isn't billed.

The eval-refund is the only thing standing between Atlas's cost claim and the long history of inference brokers promising the quantized version is fine. Bind a numeric threshold; calls below it never show up on your invoice. On every plan.

01 · Bind

Bind an evaluator with a min_score to your operation.

Regex, LLM-judge, structural — any evaluator that returns a numeric score. Set once per operation in /console/evaluators.

02 · Score

Atlas runs the evaluator on the served output, on the call path.

Same after() hook that meters usage. Score appears on the call record within about a second.

03 · Refund

Score below min_score → call's net metered quantity is zero.

A compensating Stripe meter event fires automatically. The call still appears in /console/calls so your team can correct it via feedback — but it never appears on your invoice.

Works on PAYG. Works on Reliability Loop. Works on Strategic. The mechanic is platform-wide — the only requirement is a bound evaluator with a numeric threshold and metadata.operation_id on the call.

Plans

What each plan unlocks.

Usage rates are the same across all plans — the per-call tier (Realtime / Standard / Batch) decides the bill. Plan choice gates the reliability loop, the quality refund, and Atlas Network defaults.

Pay as you go

$5 free credits · free models, no card

Match-or-better rate on every provider. We never charge more than direct would have, and usually less because Atlas routes to a cheaper variant when quality holds. Start free on free models with no card; add one for the full catalogue. Thumbs-down any call you don't like and we credit it back automatically.

·Free OpenRouter models — no card required
·$5 free credits when you add a card
·No markup — pay what the provider would charge, or less
·Auto top-up keeps you running ($25 when balance < $5; configurable)
·Thumbs-down any call → automatic credit-back
·OpenAI + Anthropic wire formats supported
·Drop-in for Codex CLI, Claude Code, Cursor, langchain, llamaindex

Start free

Reliability Loop

$1,500 / mo

platform fee + 10% on usage

Everything in Pay as you go plus the productized reliability workflow — operation-aware routing, eval-gated auto-refund, evaluators as a first-class concept, ship gates as a deploy concept, dataset promotion, and per-tenant LoRA training. The 10% markup pays for the loop; the savings on the routing side usually cover it.

·Everything in Pay as you go
·operation_id-aware routing (your prior eval scores drive picks)
·Eval-gated auto-refund (binds an evaluator + min_score per op)
·Per-tenant LoRA adapters trained on your traffic
·Atlas Network routing ON by default (override per call / org)
·30-day bill-cut guarantee — platform fee refunded if bill isn't down 30%
·99.9% uptime SLO + 1-business-day support

Talk to sales

Strategic

Custom

annual commitment

For teams running significant volume. Negotiated rates per quantization tier, dedicated capacity, and a solutions engineer who knows your evals.

·Negotiated rates per tier × quantization × provider
·Everything in Reliability Loop
·Per-operation routing policies
·Dedicated LoRA training capacity
·99.95% SLO + dedicated capacity
·Embedded solutions engineer

Contact sales

On Pay as you go, thumbs-down any call and we credit it back automatically (fair-use limits in the terms). On the Reliability Loop tier this stacks with the eval-gated auto-refund: bind an evaluator with a numeric min_score to any operation, and calls below threshold are refunded synchronously without anyone hitting thumbs-down. Strategic customers negotiate per-tier rates and dedicated capacity.

Available models

Any model. One key.

Pass any model id in the model field. The most-used models across the network this week are pinned to the top; the full provider-grouped catalogue follows. Pay as you go is at-or-below the upstream provider's list price — no markup. Reliability Loop carries the standard 10% itemized separately.

Tier

321 models supported

Model ID	Context	Realtime	Standard	Batch	Atlas-Net
Most popular this week · top 12
#1deepseek/deepseek-v4-flash DeepSeek: DeepSeek V4 Flash	1.0M	$0.1 / 1M	$0.1 / 1Mq4 · DeepInfra	$0.1 / 1M	soon
#2tencent/hy3-preview Tencent: Hy3 preview	262K	$0.066 / 1M	—	—	—
#3anthropic/claude-opus-4.7 Anthropic: Claude Opus 4.7	1M	$5.00 / 1M	—	—	—
#4anthropic/claude-sonnet-4.6 Anthropic: Claude Sonnet 4.6	1M	$3.00 / 1M	—	—	—
#5deepseek/deepseek-v4-pro DeepSeek: DeepSeek V4 Pro	1.0M	$0.435 / 1M	$1.30 / 1Mq4 · DeepInfra	$1.30 / 1M	soon
#6google/gemini-3-flash-preview Google: Gemini 3 Flash Preview	1.0M	$0.5 / 1M	—	—	—
#7deepseek/deepseek-v3.2 DeepSeek: DeepSeek V3.2	131K	$0.252 / 1M	$0.252 / 1Mq8 · Baidu	$0.252 / 1M	soon
#8nvidia/nemotron-3-super-120b-a12b NVIDIA: Nemotron 3 Super	1M	$0.09 / 1M	$0.09 / 1Mq8 · DekaLLM	$0.09 / 1M	soon
#9google/gemini-2.5-flash-lite Google: Gemini 2.5 Flash Lite	1.0M	$0.1 / 1M	—	—	—
#10moonshotai/kimi-k2.6 MoonshotAI: Kimi K2.6	262K	$0.73 / 1M	$0.73 / 1Mq4 · Io Net	$0.73 / 1M	soon
#11google/gemini-2.5-flash Google: Gemini 2.5 Flash	1.0M	$0.3 / 1M	—	—	—
#12minimax/minimax-m2.7 MiniMax: MiniMax M2.7	205K	$0.279 / 1M	$0.3 / 1Mq8 · Minimax	$0.3 / 1M	soon
OpenAI · 63 models
openai/gpt-audio OpenAI: GPT Audio	128K	$2.50 / 1M	—	—	—
openai/gpt-audio-mini OpenAI: GPT Audio Mini	128K	$0.6 / 1M	—	—	—
openai/gpt-chat-latest OpenAI: GPT Chat Latest	400K	$5.00 / 1M	—	—	—
openai/gpt-3.5-turbo OpenAI: GPT-3.5 Turbo	16K	$0.5 / 1M	—	—	—
openai/gpt-3.5-turbo-0613 OpenAI: GPT-3.5 Turbo (older v0613)	4K	$1.00 / 1M	—	—	—
openai/gpt-3.5-turbo-16k OpenAI: GPT-3.5 Turbo 16k	16K	$3.00 / 1M	—	—	—
openai/gpt-3.5-turbo-instruct OpenAI: GPT-3.5 Turbo Instruct	4K	$1.50 / 1M	—	—	—
openai/gpt-4 OpenAI: GPT-4	8K	$30.00 / 1M	—	—	—
openai/gpt-4-0314 OpenAI: GPT-4 (older v0314)	8K	$30.00 / 1M	—	—	—
openai/gpt-4-turbo OpenAI: GPT-4 Turbo	128K	$10.00 / 1M	—	—	—
openai/gpt-4-1106-preview OpenAI: GPT-4 Turbo (older v1106)	128K	$10.00 / 1M	—	—	—
openai/gpt-4-turbo-preview OpenAI: GPT-4 Turbo Preview	128K	$10.00 / 1M	—	—	—
openai/gpt-4.1 OpenAI: GPT-4.1	1.0M	$2.00 / 1M	—	—	—
openai/gpt-4.1-mini OpenAI: GPT-4.1 Mini	1.0M	$0.4 / 1M	—	—	—
openai/gpt-4.1-nano OpenAI: GPT-4.1 Nano	1.0M	$0.1 / 1M	—	—	—
openai/gpt-4o OpenAI: GPT-4o	128K	$2.50 / 1M	—	—	—
openai/gpt-4o-2024-05-13 OpenAI: GPT-4o (2024-05-13)	128K	$5.00 / 1M	—	—	—
openai/gpt-4o-2024-08-06 OpenAI: GPT-4o (2024-08-06)	128K	$2.50 / 1M	—	—	—
openai/gpt-4o-2024-11-20 OpenAI: GPT-4o (2024-11-20)	128K	$2.50 / 1M	—	—	—
openai/gpt-4o-audio-preview OpenAI: GPT-4o Audio	128K	$2.50 / 1M	—	—	—
openai/gpt-4o-search-preview OpenAI: GPT-4o Search Preview	128K	$2.50 / 1M	—	—	—
openai/gpt-4o-mini OpenAI: GPT-4o-mini	128K	$0.15 / 1M	—	—	—
openai/gpt-4o-mini-2024-07-18 OpenAI: GPT-4o-mini (2024-07-18)	128K	$0.15 / 1M	—	—	—
openai/gpt-4o-mini-search-preview OpenAI: GPT-4o-mini Search Preview	128K	$0.15 / 1M	—	—	—
openai/gpt-5 OpenAI: GPT-5	400K	$1.25 / 1M	—	—	—
openai/gpt-5-chat OpenAI: GPT-5 Chat	128K	$1.25 / 1M	—	—	—
openai/gpt-5-codex OpenAI: GPT-5 Codex	400K	$1.25 / 1M	—	—	—
openai/gpt-5-image OpenAI: GPT-5 Image	400K	$10.00 / 1M	—	—	—
openai/gpt-5-image-mini OpenAI: GPT-5 Image Mini	400K	$2.50 / 1M	—	—	—
openai/gpt-5-mini OpenAI: GPT-5 Mini	400K	$0.25 / 1M	—	—	—
openai/gpt-5-nano OpenAI: GPT-5 Nano	400K	$0.05 / 1M	—	—	—
openai/gpt-5-pro OpenAI: GPT-5 Pro	400K	$15.00 / 1M	—	—	—
openai/gpt-5.1 OpenAI: GPT-5.1	400K	$1.25 / 1M	—	—	—
openai/gpt-5.1-chat OpenAI: GPT-5.1 Chat	128K	$1.25 / 1M	—	—	—
openai/gpt-5.1-codex OpenAI: GPT-5.1-Codex	400K	$1.25 / 1M	—	—	—
openai/gpt-5.1-codex-max OpenAI: GPT-5.1-Codex-Max	400K	$1.25 / 1M	—	—	—
openai/gpt-5.1-codex-mini OpenAI: GPT-5.1-Codex-Mini	400K	$0.25 / 1M	—	—	—
openai/gpt-5.2 OpenAI: GPT-5.2	400K	$1.75 / 1M	—	—	—
openai/gpt-5.2-chat OpenAI: GPT-5.2 Chat	128K	$1.75 / 1M	—	—	—
openai/gpt-5.2-pro OpenAI: GPT-5.2 Pro	400K	$21.00 / 1M	—	—	—
openai/gpt-5.2-codex OpenAI: GPT-5.2-Codex	400K	$1.75 / 1M	—	—	—
openai/gpt-5.3-chat OpenAI: GPT-5.3 Chat	128K	$1.75 / 1M	—	—	—
openai/gpt-5.3-codex OpenAI: GPT-5.3-Codex	400K	$1.75 / 1M	—	—	—
openai/gpt-5.4 OpenAI: GPT-5.4	1.1M	$2.50 / 1M	—	—	—
openai/gpt-5.4-image-2 OpenAI: GPT-5.4 Image 2	272K	$8.00 / 1M	—	—	—
openai/gpt-5.4-mini OpenAI: GPT-5.4 Mini	400K	$0.75 / 1M	—	—	—
openai/gpt-5.4-nano OpenAI: GPT-5.4 Nano	400K	$0.2 / 1M	—	—	—
openai/gpt-5.4-pro OpenAI: GPT-5.4 Pro	1.1M	$30.00 / 1M	—	—	—
openai/gpt-5.5 OpenAI: GPT-5.5	1.1M	$5.00 / 1M	—	—	—
openai/gpt-5.5-pro OpenAI: GPT-5.5 Pro	1.1M	$30.00 / 1M	—	—	—
openai/gpt-oss-120b OpenAI: gpt-oss-120b	131K	$0.039 / 1M	$0.05 / 1Mq4 · Novita	$0.05 / 1M	—
openai/gpt-oss-20b OpenAI: gpt-oss-20b	131K	$0.03 / 1M	$0.04 / 1Mq4 · Novita	$0.04 / 1M	—
openai/gpt-oss-safeguard-20b OpenAI: gpt-oss-safeguard-20b	131K	$0.075 / 1M	—	—	—
openai/o1 OpenAI: o1	200K	$15.00 / 1M	—	—	—
openai/o1-pro OpenAI: o1-pro	200K	$150.00 / 1M	—	—	—
openai/o3 OpenAI: o3	200K	$2.00 / 1M	—	—	—
openai/o3-deep-research OpenAI: o3 Deep Research	200K	$10.00 / 1M	—	—	—
openai/o3-mini OpenAI: o3 Mini	200K	$1.10 / 1M	—	—	—
openai/o3-mini-high OpenAI: o3 Mini High	200K	$1.10 / 1M	—	—	—
openai/o3-pro OpenAI: o3 Pro	200K	$20.00 / 1M	—	—	—
openai/o4-mini OpenAI: o4 Mini	200K	$1.10 / 1M	—	—	—
openai/o4-mini-deep-research OpenAI: o4 Mini Deep Research	200K	$2.00 / 1M	—	—	—
openai/o4-mini-high OpenAI: o4 Mini High	200K	$1.10 / 1M	—	—	—
Anthropic · 13 models
anthropic/claude-3-haiku Anthropic: Claude 3 Haiku	200K	$0.25 / 1M	—	—	—
anthropic/claude-3.5-haiku Anthropic: Claude 3.5 Haiku	200K	$0.8 / 1M	—	—	—
anthropic/claude-haiku-4.5 Anthropic: Claude Haiku 4.5	200K	$1.00 / 1M	—	—	—
anthropic/claude-opus-4 Anthropic: Claude Opus 4	200K	$15.00 / 1M	—	—	—
anthropic/claude-opus-4.1 Anthropic: Claude Opus 4.1	200K	$15.00 / 1M	—	—	—
anthropic/claude-opus-4.5 Anthropic: Claude Opus 4.5	200K	$5.00 / 1M	—	—	—
anthropic/claude-opus-4.6 Anthropic: Claude Opus 4.6	1M	$5.00 / 1M	—	—	—
anthropic/claude-opus-4.6-fast Anthropic: Claude Opus 4.6 (Fast)	1M	$30.00 / 1M	—	—	—
anthropic/claude-opus-4.7 Anthropic: Claude Opus 4.7	1M	$5.00 / 1M	—	—	—
anthropic/claude-opus-4.7-fast Anthropic: Claude Opus 4.7 (Fast)	1M	$30.00 / 1M	—	—	—
anthropic/claude-sonnet-4 Anthropic: Claude Sonnet 4	1M	$3.00 / 1M	—	—	—
anthropic/claude-sonnet-4.5 Anthropic: Claude Sonnet 4.5	1M	$3.00 / 1M	—	—	—
anthropic/claude-sonnet-4.6 Anthropic: Claude Sonnet 4.6	1M	$3.00 / 1M	—	—	—
Google · 24 models
google/gemini-2.0-flash-001 Google: Gemini 2.0 Flash	1M	$0.1 / 1M	—	—	—
google/gemini-2.0-flash-lite-001 Google: Gemini 2.0 Flash Lite	1.0M	$0.075 / 1M	—	—	—
google/gemini-2.5-flash Google: Gemini 2.5 Flash	1.0M	$0.3 / 1M	—	—	—
google/gemini-2.5-flash-lite Google: Gemini 2.5 Flash Lite	1.0M	$0.1 / 1M	—	—	—
google/gemini-2.5-flash-lite-preview-09-2025 Google: Gemini 2.5 Flash Lite Preview 09-2025	1.0M	$0.1 / 1M	—	—	—
google/gemini-2.5-pro Google: Gemini 2.5 Pro	1.0M	$1.25 / 1M	—	—	—
google/gemini-2.5-pro-preview-05-06 Google: Gemini 2.5 Pro Preview 05-06	1.0M	$1.25 / 1M	—	—	—
google/gemini-2.5-pro-preview Google: Gemini 2.5 Pro Preview 06-05	1.0M	$1.25 / 1M	—	—	—
google/gemini-3-flash-preview Google: Gemini 3 Flash Preview	1.0M	$0.5 / 1M	—	—	—
google/gemini-3.1-flash-lite Google: Gemini 3.1 Flash Lite	1.0M	$0.25 / 1M	—	—	—
google/gemini-3.1-flash-lite-preview Google: Gemini 3.1 Flash Lite Preview	1.0M	$0.25 / 1M	—	—	—
google/gemini-3.1-pro-preview Google: Gemini 3.1 Pro Preview	1.0M	$2.00 / 1M	—	—	—
google/gemini-3.1-pro-preview-customtools Google: Gemini 3.1 Pro Preview Custom Tools	1.0M	$2.00 / 1M	—	—	—
google/gemini-3.5-flash Google: Gemini 3.5 Flash	1.0M	$1.50 / 1M	—	—	—
google/gemma-2-27b-it Google: Gemma 2 27B	8K	$0.65 / 1M	$0.65 / 1Mq4 · NextBit	$0.65 / 1M	soon
google/gemma-3-12b-it Google: Gemma 3 12B	131K	$0.04 / 1M	—	—	—
google/gemma-3-27b-it Google: Gemma 3 27B	131K	$0.08 / 1M	$0.08 / 1Mq8 · DeepInfra	$0.08 / 1M	soon
google/gemma-3-4b-it Google: Gemma 3 4B	131K	$0.04 / 1M	—	—	—
google/gemma-3n-e4b-it Google: Gemma 3n 4B	33K	$0.06 / 1M	—	—	—
google/gemma-4-26b-a4b-it Google: Gemma 4 26B A4B	262K	$0.06 / 1M	$0.07 / 1Mq8 · DeepInfra	$0.07 / 1M	soon
google/gemma-4-31b-it Google: Gemma 4 31B	262K	$0.12 / 1M	$0.12 / 1Mq4 · DeepInfra	$0.12 / 1M	soon
google/gemini-2.5-flash-image Google: Nano Banana (Gemini 2.5 Flash Image)	33K	$0.3 / 1M	—	—	—
google/gemini-3.1-flash-image-preview Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview)	131K	$0.5 / 1M	—	—	—
google/gemini-3-pro-image-preview Google: Nano Banana Pro (Gemini 3 Pro Image Preview)	66K	$2.00 / 1M	—	—	—
Meta · 12 models
meta-llama/llama-guard-3-8b Llama Guard 3 8B	131K	$0.484 / 1M	—	—	—
meta-llama/llama-3-70b-instruct Meta: Llama 3 70B Instruct	8K	$0.51 / 1M	$0.51 / 1Mq8 · Novita	$0.51 / 1M	soon
meta-llama/llama-3-8b-instruct Meta: Llama 3 8B Instruct	8K	$0.04 / 1M	$0.1 / 1Mq4 · Together	$0.1 / 1M	soon
meta-llama/llama-3.1-70b-instruct Meta: Llama 3.1 70B Instruct	131K	$0.4 / 1M	$0.4 / 1Mq8 · DeepInfra	$0.4 / 1M	soon
meta-llama/llama-3.1-8b-instruct Meta: Llama 3.1 8B Instruct	131K	$0.02 / 1M	$0.02 / 1Mq8 · Novita	$0.02 / 1M	soon
meta-llama/llama-3.2-11b-vision-instruct Meta: Llama 3.2 11B Vision Instruct	131K	$0.245 / 1M	$0.245 / 1Mq8 · DeepInfra	$0.245 / 1M	soon
meta-llama/llama-3.2-1b-instruct Meta: Llama 3.2 1B Instruct	131K	$0.027 / 1M	—	—	—
meta-llama/llama-3.2-3b-instruct Meta: Llama 3.2 3B Instruct	131K	$0.051 / 1M	—	—	—
meta-llama/llama-3.3-70b-instruct Meta: Llama 3.3 70B Instruct	131K	$0.1 / 1M	$0.1 / 1Mq8 · DeepInfra	$0.1 / 1M	soon
meta-llama/llama-4-maverick Meta: Llama 4 Maverick	1.0M	$0.15 / 1M	$0.15 / 1Mq8 · DeepInfra	$0.15 / 1M	soon
meta-llama/llama-4-scout Meta: Llama 4 Scout	10M	$0.08 / 1M	$0.08 / 1Mq8 · DeepInfra	$0.08 / 1M	soon
meta-llama/llama-guard-4-12b Meta: Llama Guard 4 12B	164K	$0.18 / 1M	—	—	—
Mistral · 24 models
mistralai/mistral-large Mistral Large	128K	$2.00 / 1M	—	—	—
mistralai/mistral-large-2407 Mistral Large 2407	131K	$2.00 / 1M	—	—	—
mistralai/mistral-large-2411 Mistral Large 2411	131K	$2.00 / 1M	—	—	—
mistralai/codestral-2508 Mistral: Codestral 2508	256K	$0.3 / 1M	—	—	—
mistralai/devstral-2512 Mistral: Devstral 2 2512	262K	$0.4 / 1M	—	—	—
mistralai/devstral-medium Mistral: Devstral Medium	131K	$0.4 / 1M	—	—	—
mistralai/devstral-small Mistral: Devstral Small 1.1	131K	$0.1 / 1M	—	—	—
mistralai/ministral-14b-2512 Mistral: Ministral 3 14B 2512	262K	$0.2 / 1M	$0.35 / 1Mq8 · NextBit	$0.35 / 1M	soon
mistralai/ministral-3b-2512 Mistral: Ministral 3 3B 2512	131K	$0.1 / 1M	$0.15 / 1Mq8 · NextBit	$0.15 / 1M	soon
mistralai/ministral-8b-2512 Mistral: Ministral 3 8B 2512	262K	$0.15 / 1M	$0.3 / 1Mq8 · NextBit	$0.3 / 1M	soon
mistralai/mistral-7b-instruct-v0.1 Mistral: Mistral 7B Instruct v0.1	4K	$0.11 / 1M	—	—	—
mistralai/mistral-large-2512 Mistral: Mistral Large 3 2512	262K	$0.5 / 1M	—	—	—
mistralai/mistral-medium-3 Mistral: Mistral Medium 3	131K	$0.4 / 1M	—	—	—
mistralai/mistral-medium-3.1 Mistral: Mistral Medium 3.1	131K	$0.4 / 1M	—	—	—
mistralai/mistral-medium-3-5 Mistral: Mistral Medium 3.5	262K	$1.50 / 1M	—	—	—
mistralai/mistral-nemo Mistral: Mistral Nemo	131K	$0.02 / 1M	$0.02 / 1Mq8 · DeepInfra	$0.02 / 1M	soon
mistralai/mistral-small-24b-instruct-2501 Mistral: Mistral Small 3	33K	$0.05 / 1M	$0.05 / 1Mq8 · DeepInfra	$0.05 / 1M	soon
mistralai/mistral-small-3.1-24b-instruct Mistral: Mistral Small 3.1 24B	128K	$0.351 / 1M	—	—	—
mistralai/mistral-small-3.2-24b-instruct Mistral: Mistral Small 3.2 24B	128K	$0.075 / 1M	$0.075 / 1Mq8 · DeepInfra	$0.075 / 1M	soon
mistralai/mistral-small-2603 Mistral: Mistral Small 4	262K	$0.15 / 1M	$0.188 / 1Mq8 · Venice	$0.188 / 1M	soon
mistralai/mixtral-8x22b-instruct Mistral: Mixtral 8x22B Instruct	66K	$2.00 / 1M	—	—	—
mistralai/pixtral-large-2411 Mistral: Pixtral Large 2411	131K	$2.00 / 1M	—	—	—
mistralai/mistral-saba Mistral: Saba	33K	$0.2 / 1M	—	—	—
mistralai/voxtral-small-24b-2507 Mistral: Voxtral Small 24B 2507	32K	$0.1 / 1M	—	—	—
xAI · 4 models
x-ai/grok-4.20 xAI: Grok 4.20	2M	$1.25 / 1M	—	—	—
x-ai/grok-4.20-multi-agent xAI: Grok 4.20 Multi-Agent	2M	$2.00 / 1M	—	—	—
x-ai/grok-4.3 xAI: Grok 4.3	1M	$1.25 / 1M	—	—	—
x-ai/grok-build-0.1 xAI: Grok Build 0.1	256K	$1.00 / 1M	—	—	—
DeepSeek · 13 models
deepseek/deepseek-chat DeepSeek: DeepSeek V3	131K	$0.229 / 1M	$0.32 / 1Mq4 · DeepInfra	$0.32 / 1M	soon
deepseek/deepseek-chat-v3-0324 DeepSeek: DeepSeek V3 0324	164K	$0.2 / 1M	$0.2 / 1Mq4 · DeepInfra	$0.2 / 1M	soon
deepseek/deepseek-chat-v3.1 DeepSeek: DeepSeek V3.1	164K	$0.21 / 1M	$0.21 / 1Mq4 · DeepInfra	$0.21 / 1M	soon
deepseek/deepseek-v3.1-terminus DeepSeek: DeepSeek V3.1 Terminus	164K	$0.27 / 1M	$0.27 / 1Mq4 · DeepInfra	$0.27 / 1M	soon
deepseek/deepseek-v3.2 DeepSeek: DeepSeek V3.2	131K	$0.252 / 1M	$0.252 / 1Mq8 · Baidu	$0.252 / 1M	soon
deepseek/deepseek-v3.2-exp DeepSeek: DeepSeek V3.2 Exp	164K	$0.27 / 1M	$0.27 / 1Mq8 · AtlasCloud	$0.27 / 1M	soon
deepseek/deepseek-v3.2-speciale DeepSeek: DeepSeek V3.2 Speciale	164K	$0.287 / 1M	$0.287 / 1Mq8 · AtlasCloud	$0.287 / 1M	soon
deepseek/deepseek-v4-flash DeepSeek: DeepSeek V4 Flash	1.0M	$0.1 / 1M	$0.1 / 1Mq4 · DeepInfra	$0.1 / 1M	soon
deepseek/deepseek-v4-pro DeepSeek: DeepSeek V4 Pro	1.0M	$0.435 / 1M	$1.30 / 1Mq4 · DeepInfra	$1.30 / 1M	soon
deepseek/deepseek-r1 DeepSeek: R1	164K	$0.7 / 1M	$0.7 / 1Mq8 · Novita	$0.7 / 1M	soon
deepseek/deepseek-r1-0528 DeepSeek: R1 0528	164K	$0.5 / 1M	$0.5 / 1Mq4 · DeepInfra	$0.5 / 1M	soon
deepseek/deepseek-r1-distill-llama-70b DeepSeek: R1 Distill Llama 70B	131K	$0.7 / 1M	$0.7 / 1Mq8 · DeepInfra	$0.7 / 1M	soon
deepseek/deepseek-r1-distill-qwen-32b DeepSeek: R1 Distill Qwen 32B	128K	$0.29 / 1M	$0.29 / 1Mq8 · NextBit	$0.29 / 1M	soon
Alibaba (Qwen) · 46 models
qwen/qwen-plus-2025-07-28 Qwen: Qwen Plus 0728	1M	$0.26 / 1M	—	—	—
qwen/qwen-plus-2025-07-28:thinking Qwen: Qwen Plus 0728 (thinking)	1M	$0.26 / 1M	—	—	—
qwen/qwen-plus Qwen: Qwen-Plus	1M	$0.26 / 1M	—	—	—
qwen/qwen-2.5-7b-instruct Qwen: Qwen2.5 7B Instruct	131K	$0.04 / 1M	$0.04 / 1Mq8 · AtlasCloud	$0.04 / 1M	soon
qwen/qwen2.5-vl-72b-instruct Qwen: Qwen2.5 VL 72B Instruct	131K	$0.25 / 1M	$0.25 / 1Mq8 · Nebius	$0.25 / 1M	soon
qwen/qwen3-14b Qwen: Qwen3 14B	132K	$0.1 / 1M	$0.1 / 1Mq4 · NextBit	$0.1 / 1M	soon
qwen/qwen3-235b-a22b Qwen: Qwen3 235B A22B	131K	$0.455 / 1M	—	—	—
qwen/qwen3-235b-a22b-2507 Qwen: Qwen3 235B A22B Instruct 2507	262K	$0.071 / 1M	$0.071 / 1Mq8 · DeepInfra	$0.071 / 1M	soon
qwen/qwen3-235b-a22b-thinking-2507 Qwen: Qwen3 235B A22B Thinking 2507	262K	$0.149 / 1M	$0.23 / 1Mq8 · DeepInfra	$0.23 / 1M	soon
qwen/qwen3-30b-a3b Qwen: Qwen3 30B A3B	131K	$0.09 / 1M	$0.09 / 1Mq8 · DeepInfra	$0.09 / 1M	soon
qwen/qwen3-30b-a3b-instruct-2507 Qwen: Qwen3 30B A3B Instruct 2507	262K	$0.09 / 1M	$0.09 / 1Mq8 · SiliconFlow	$0.09 / 1M	soon
qwen/qwen3-30b-a3b-thinking-2507 Qwen: Qwen3 30B A3B Thinking 2507	131K	$0.08 / 1M	$0.08 / 1Mq8 · AtlasCloud	$0.08 / 1M	soon
qwen/qwen3-32b Qwen: Qwen3 32B	131K	$0.08 / 1M	$0.08 / 1Mq8 · DeepInfra	$0.08 / 1M	soon
qwen/qwen3-8b Qwen: Qwen3 8B	131K	$0.05 / 1M	$0.05 / 1Mq8 · AtlasCloud	$0.05 / 1M	soon
qwen/qwen3-coder-30b-a3b-instruct Qwen: Qwen3 Coder 30B A3B Instruct	160K	$0.07 / 1M	$0.07 / 1Mq8 · Novita	$0.07 / 1M	soon
qwen/qwen3-coder Qwen: Qwen3 Coder 480B A35B	1.0M	$0.22 / 1M	$0.3 / 1Mq4 · DeepInfra	$0.3 / 1M	soon
qwen/qwen3-coder-flash Qwen: Qwen3 Coder Flash	1M	$0.195 / 1M	—	—	—
qwen/qwen3-coder-next Qwen: Qwen3 Coder Next	262K	$0.11 / 1M	$0.11 / 1Mq8 · Ionstream	$0.11 / 1M	soon
qwen/qwen3-coder-plus Qwen: Qwen3 Coder Plus	1M	$0.65 / 1M	—	—	—
qwen/qwen3-max Qwen: Qwen3 Max	262K	$0.78 / 1M	—	—	—
qwen/qwen3-max-thinking Qwen: Qwen3 Max Thinking	262K	$0.78 / 1M	—	—	—
qwen/qwen3-next-80b-a3b-instruct Qwen: Qwen3 Next 80B A3B Instruct	262K	$0.09 / 1M	$0.09 / 1Mq8 · DeepInfra	$0.09 / 1M	soon
qwen/qwen3-next-80b-a3b-thinking Qwen: Qwen3 Next 80B A3B Thinking	262K	$0.098 / 1M	$0.15 / 1Mq8 · AtlasCloud	$0.15 / 1M	soon
qwen/qwen3-vl-235b-a22b-instruct Qwen: Qwen3 VL 235B A22B Instruct	262K	$0.2 / 1M	$0.2 / 1Mq8 · DeepInfra	$0.2 / 1M	soon
qwen/qwen3-vl-235b-a22b-thinking Qwen: Qwen3 VL 235B A22B Thinking	131K	$0.26 / 1M	—	—	—
qwen/qwen3-vl-30b-a3b-instruct Qwen: Qwen3 VL 30B A3B Instruct	262K	$0.13 / 1M	$0.15 / 1Mq8 · AtlasCloud	$0.15 / 1M	soon
qwen/qwen3-vl-30b-a3b-thinking Qwen: Qwen3 VL 30B A3B Thinking	131K	$0.13 / 1M	$0.29 / 1Mq8 · SiliconFlow	$0.29 / 1M	soon
qwen/qwen3-vl-32b-instruct Qwen: Qwen3 VL 32B Instruct	262K	$0.104 / 1M	—	—	—
qwen/qwen3-vl-8b-instruct Qwen: Qwen3 VL 8B Instruct	256K	$0.08 / 1M	$0.08 / 1Mq8 · AtlasCloud	$0.08 / 1M	soon
qwen/qwen3-vl-8b-thinking Qwen: Qwen3 VL 8B Thinking	256K	$0.117 / 1M	—	—	—
qwen/qwen3.5-397b-a17b Qwen: Qwen3.5 397B A17B	262K	$0.39 / 1M	$0.45 / 1Mq8 · Chutes	$0.45 / 1M	soon
qwen/qwen3.5-plus-02-15 Qwen: Qwen3.5 Plus 2026-02-15	1M	$0.26 / 1M	—	—	—
qwen/qwen3.5-plus-20260420 Qwen: Qwen3.5 Plus 2026-04-20	1M	$0.3 / 1M	—	—	—
qwen/qwen3.5-122b-a10b Qwen: Qwen3.5-122B-A10B	262K	$0.26 / 1M	$0.26 / 1Mq8 · SiliconFlow	$0.26 / 1M	soon
qwen/qwen3.5-27b Qwen: Qwen3.5-27B	262K	$0.195 / 1M	$0.25 / 1Mq8 · SiliconFlow	$0.25 / 1M	soon
qwen/qwen3.5-35b-a3b Qwen: Qwen3.5-35B-A3B	262K	$0.139 / 1M	$0.139 / 1Mq8 · DekaLLM	$0.139 / 1M	soon
qwen/qwen3.5-9b Qwen: Qwen3.5-9B	262K	$0.04 / 1M	$0.1 / 1Mq8 · SiliconFlow	$0.1 / 1M	soon
qwen/qwen3.5-flash-02-23 Qwen: Qwen3.5-Flash	1M	$0.065 / 1M	—	—	—
qwen/qwen3.6-27b Qwen: Qwen3.6 27B	262K	$0.29 / 1M	$0.29 / 1Mq8 · Io Net	$0.29 / 1M	soon
qwen/qwen3.6-35b-a3b Qwen: Qwen3.6 35B A3B	262K	$0.14 / 1M	$0.14 / 1Mq8 · Io Net	$0.14 / 1M	soon
qwen/qwen3.6-flash Qwen: Qwen3.6 Flash	1M	$0.188 / 1M	—	—	—
qwen/qwen3.6-max-preview Qwen: Qwen3.6 Max Preview	262K	$1.04 / 1M	—	—	—
qwen/qwen3.6-plus Qwen: Qwen3.6 Plus	1M	$0.325 / 1M	—	—	—
qwen/qwen3.7-max Qwen: Qwen3.7 Max	1M	$1.25 / 1M	—	—	—
qwen/qwen-2.5-72b-instruct Qwen2.5 72B Instruct	131K	$0.36 / 1M	$0.36 / 1Mq8 · DeepInfra	$0.36 / 1M	soon
qwen/qwen-2.5-coder-32b-instruct Qwen2.5 Coder 32B Instruct	128K	$0.66 / 1M	—	—	—
Cohere · 4 models
cohere/command-a Cohere: Command A	256K	$2.50 / 1M	—	—	—
cohere/command-r-08-2024 Cohere: Command R (08-2024)	128K	$0.15 / 1M	—	—	—
cohere/command-r-plus-08-2024 Cohere: Command R+ (08-2024)	128K	$2.50 / 1M	—	—	—
cohere/command-r7b-12-2024 Cohere: Command R7B (12-2024)	128K	$0.037 / 1M	—	—	—
Amazon · 5 models
amazon/nova-2-lite-v1 Amazon: Nova 2 Lite	1M	$0.3 / 1M	—	—	—
amazon/nova-lite-v1 Amazon: Nova Lite 1.0	300K	$0.06 / 1M	—	—	—
amazon/nova-micro-v1 Amazon: Nova Micro 1.0	128K	$0.035 / 1M	—	—	—
amazon/nova-premier-v1 Amazon: Nova Premier 1.0	1M	$2.50 / 1M	—	—	—
amazon/nova-pro-v1 Amazon: Nova Pro 1.0	300K	$0.8 / 1M	—	—	—
Microsoft · 3 models
microsoft/phi-4 Microsoft: Phi 4	16K	$0.065 / 1M	$0.065 / 1Mq4 · NextBit	$0.065 / 1M	soon
microsoft/phi-4-mini-instruct Microsoft: Phi 4 Mini Instruct	131K	$0.08 / 1M	—	—	—
microsoft/wizardlm-2-8x22b WizardLM-2 8x22B	66K	$0.62 / 1M	—	—	—
NVIDIA · 4 models
nvidia/llama-3.3-nemotron-super-49b-v1.5 NVIDIA: Llama 3.3 Nemotron Super 49B V1.5	131K	$0.1 / 1M	$0.1 / 1Mq8 · DeepInfra	$0.1 / 1M	soon
nvidia/nemotron-3-nano-30b-a3b NVIDIA: Nemotron 3 Nano 30B A3B	262K	$0.05 / 1M	$0.05 / 1Mq4 · DeepInfra	$0.05 / 1M	soon
nvidia/nemotron-3-super-120b-a12b NVIDIA: Nemotron 3 Super	1M	$0.09 / 1M	$0.09 / 1Mq8 · DekaLLM	$0.09 / 1M	soon
nvidia/nemotron-nano-9b-v2 NVIDIA: Nemotron Nano 9B V2	131K	$0.04 / 1M	—	—	—
Perplexity · 5 models
perplexity/sonar Perplexity: Sonar	127K	$1.00 / 1M	—	—	—
perplexity/sonar-deep-research Perplexity: Sonar Deep Research	128K	$2.00 / 1M	—	—	—
perplexity/sonar-pro Perplexity: Sonar Pro	200K	$3.00 / 1M	—	—	—
perplexity/sonar-pro-search Perplexity: Sonar Pro Search	200K	$3.00 / 1M	—	—	—
perplexity/sonar-reasoning-pro Perplexity: Sonar Reasoning Pro	128K	$2.00 / 1M	—	—	—
AI21 · 1 model
ai21/jamba-large-1.7 AI21: Jamba Large 1.7	256K	$2.00 / 1M	$2.00 / 1Mq8 · AI21	$2.00 / 1M	—
Aion · 4 models
aion-labs/aion-1.0 AionLabs: Aion-1.0	131K	$4.00 / 1M	—	—	—
aion-labs/aion-1.0-mini AionLabs: Aion-1.0-Mini	131K	$0.7 / 1M	—	—	—
aion-labs/aion-2.0 AionLabs: Aion-2.0	131K	$0.8 / 1M	—	—	—
aion-labs/aion-rp-llama-3.1-8b AionLabs: Aion-RP 1.0 (8B)	33K	$0.8 / 1M	—	—	—
AlfredPros · 1 model
alfredpros/codellama-7b-instruct-solidity AlfredPros: CodeLLaMa 7B Instruct Solidity	4K	$0.8 / 1M	—	—	—
Allen AI · 1 model
allenai/olmo-3-32b-think AllenAI: Olmo 3 32B Think	66K	$0.15 / 1M	—	—	—
Anthracite · 1 model
anthracite-org/magnum-v4-72b Magnum v4 72B	33K	$3.00 / 1M	$3.00 / 1Mq8 · Mancer 2	$3.00 / 1M	soon
Arcee · 6 models
arcee-ai/coder-large Arcee AI: Coder Large	33K	$0.5 / 1M	—	—	—
arcee-ai/maestro-reasoning Arcee AI: Maestro Reasoning	131K	$0.9 / 1M	—	—	—
arcee-ai/spotlight Arcee AI: Spotlight	131K	$0.18 / 1M	—	—	—
arcee-ai/trinity-large-thinking Arcee AI: Trinity Large Thinking	262K	$0.22 / 1M	$0.22 / 1Mq8 · Parasail	$0.22 / 1M	soon
arcee-ai/trinity-mini Arcee AI: Trinity Mini	131K	$0.045 / 1M	—	—	—
arcee-ai/virtuoso-large Arcee AI: Virtuoso Large	131K	$0.75 / 1M	—	—	—
Baidu · 6 models
baidu/ernie-4.5-21b-a3b Baidu: ERNIE 4.5 21B A3B	131K	$0.07 / 1M	—	—	—
baidu/ernie-4.5-21b-a3b-thinking Baidu: ERNIE 4.5 21B A3B Thinking	131K	$0.07 / 1M	$0.07 / 1Mq8 · Novita	$0.07 / 1M	soon
baidu/ernie-4.5-300b-a47b Baidu: ERNIE 4.5 300B A47B	131K	$0.28 / 1M	—	—	—
baidu/ernie-4.5-vl-28b-a3b Baidu: ERNIE 4.5 VL 28B A3B	131K	$0.14 / 1M	—	—	—
baidu/ernie-4.5-vl-424b-a47b Baidu: ERNIE 4.5 VL 424B A47B	131K	$0.42 / 1M	—	—	—
baidu/qianfan-ocr-fast Baidu: Qianfan-OCR-Fast	66K	$0.68 / 1M	$0.68 / 1Mq8 · Baidu	$0.68 / 1M	soon
ByteDance · 1 model
bytedance/ui-tars-1.5-7b ByteDance: UI-TARS 7B	128K	$0.1 / 1M	—	—	—
ByteDance Seed · 4 models
bytedance-seed/seed-1.6 ByteDance Seed: Seed 1.6	262K	$0.25 / 1M	$0.25 / 1Mq8 · Seed	$0.25 / 1M	soon
bytedance-seed/seed-1.6-flash ByteDance Seed: Seed 1.6 Flash	262K	$0.075 / 1M	$0.075 / 1Mq8 · Seed	$0.075 / 1M	soon
bytedance-seed/seed-2.0-lite ByteDance Seed: Seed-2.0-Lite	262K	$0.25 / 1M	$0.25 / 1Mq8 · Seed	$0.25 / 1M	soon
bytedance-seed/seed-2.0-mini ByteDance Seed: Seed-2.0-Mini	262K	$0.1 / 1M	$0.1 / 1Mq8 · Seed	$0.1 / 1M	soon
DeepCogito · 1 model
deepcogito/cogito-v2.1-671b Deep Cogito: Cogito v2.1 671B	128K	$1.25 / 1M	—	—	—
Essential AI · 1 model
essentialai/rnj-1-instruct EssentialAI: Rnj 1 Instruct	33K	$0.15 / 1M	—	—	—
Gryphe · 1 model
gryphe/mythomax-l2-13b MythoMax 13B	4K	$0.06 / 1M	$0.06 / 1Mq4 · NextBit	$0.06 / 1M	soon
IBM Granite · 2 models
ibm-granite/granite-4.0-h-micro IBM: Granite 4.0 Micro	131K	$0.017 / 1M	—	—	—
ibm-granite/granite-4.1-8b IBM: Granite 4.1 8B	131K	$0.05 / 1M	—	—	—
Inception · 1 model
inception/mercury-2 Inception: Mercury 2	128K	$0.25 / 1M	—	—	—
Inclusion AI · 3 models
inclusionai/ling-2.6-1t inclusionAI: Ling-2.6-1T	262K	$0.075 / 1M	—	—	—
inclusionai/ling-2.6-flash inclusionAI: Ling-2.6-flash	262K	$0.01 / 1M	—	—	—
inclusionai/ring-2.6-1t inclusionAI: Ring-2.6-1T	262K	$0.075 / 1M	—	—	—
Inflection · 2 models
inflection/inflection-3-pi Inflection: Inflection 3 Pi	8K	$2.50 / 1M	—	—	—
inflection/inflection-3-productivity Inflection: Inflection 3 Productivity	8K	$2.50 / 1M	—	—	—
Kwai · 1 model
kwaipilot/kat-coder-pro-v2 Kwaipilot: KAT-Coder-Pro V2	256K	$0.3 / 1M	$0.3 / 1Mq8 · AtlasCloud	$0.3 / 1M	soon
Liquid · 1 model
liquid/lfm-2-24b-a2b LiquidAI: LFM2-24B-A2B	128K	$0.03 / 1M	—	—	—
Mancer · 1 model
mancer/weaver Mancer: Weaver (alpha)	8K	$0.75 / 1M	$0.75 / 1Mq8 · Mancer 2	$0.75 / 1M	soon
MiniMax · 7 models
minimax/minimax-m1 MiniMax: MiniMax M1	1M	$0.4 / 1M	—	—	—
minimax/minimax-m2 MiniMax: MiniMax M2	205K	$0.255 / 1M	$0.255 / 1Mq8 · AtlasCloud	$0.255 / 1M	soon
minimax/minimax-m2-her MiniMax: MiniMax M2-her	66K	$0.3 / 1M	—	—	—
minimax/minimax-m2.1 MiniMax: MiniMax M2.1	205K	$0.29 / 1M	$0.29 / 1Mq8 · AtlasCloud	$0.29 / 1M	soon
minimax/minimax-m2.5 MiniMax: MiniMax M2.5	205K	$0.15 / 1M	$0.15 / 1Mq8 · AkashML	$0.15 / 1M	soon
minimax/minimax-m2.7 MiniMax: MiniMax M2.7	205K	$0.279 / 1M	$0.3 / 1Mq8 · Minimax	$0.3 / 1M	soon
minimax/minimax-01 MiniMax: MiniMax-01	1.0M	$0.2 / 1M	—	—	—
Moonshot · 5 models
moonshotai/kimi-k2 MoonshotAI: Kimi K2 0711	131K	$0.57 / 1M	$0.57 / 1Mq8 · Novita	$0.57 / 1M	soon
moonshotai/kimi-k2-0905 MoonshotAI: Kimi K2 0905	262K	$0.6 / 1M	$0.6 / 1Mq8 · AtlasCloud	$0.6 / 1M	soon
moonshotai/kimi-k2-thinking MoonshotAI: Kimi K2 Thinking	262K	$0.6 / 1M	$0.6 / 1Mq4 · AtlasCloud	$0.6 / 1M	soon
moonshotai/kimi-k2.5 MoonshotAI: Kimi K2.5	262K	$0.4 / 1M	$0.4 / 1Mq4 · ModelRun	$0.4 / 1M	soon
moonshotai/kimi-k2.6 MoonshotAI: Kimi K2.6	262K	$0.73 / 1M	$0.73 / 1Mq4 · Io Net	$0.73 / 1M	soon
Morph · 2 models
morph/morph-v3-fast Morph: Morph V3 Fast	82K	$0.8 / 1M	—	—	—
morph/morph-v3-large Morph: Morph V3 Large	262K	$0.9 / 1M	—	—	—
Newmen · 1 model
atlas-1 Routes to best model per call. Trains LoRA adapters per operation.	1M	Newmen rate	Newmen rateatlas picks the variant	Newmen rate	router
Nex AGI · 1 model
nex-agi/deepseek-v3.1-nex-n1 Nex AGI: DeepSeek V3.1 Nex N1	131K	$0.135 / 1M	$0.135 / 1Mq8 · SiliconFlow	$0.135 / 1M	—
Nous Research · 5 models
nousresearch/hermes-3-llama-3.1-405b Nous: Hermes 3 405B Instruct	131K	$1.00 / 1M	$1.00 / 1Mq8 · DeepInfra	$1.00 / 1M	soon
nousresearch/hermes-3-llama-3.1-70b Nous: Hermes 3 70B Instruct	131K	$0.3 / 1M	$0.3 / 1Mq8 · DeepInfra	$0.3 / 1M	soon
nousresearch/hermes-4-405b Nous: Hermes 4 405B	131K	$1.00 / 1M	$1.00 / 1Mq8 · Nebius	$1.00 / 1M	soon
nousresearch/hermes-4-70b Nous: Hermes 4 70B	131K	$0.13 / 1M	$0.13 / 1Mq8 · Nebius	$0.13 / 1M	soon
nousresearch/hermes-2-pro-llama-3-8b NousResearch: Hermes 2 Pro - Llama-3 8B	8K	$0.14 / 1M	—	—	—
OpenRouter · 3 models
openrouter/auto Auto Router	2M	$-1000000 / 1M	—	—	—
openrouter/bodybuilder Body Builder (beta)	128K	$-1000000 / 1M	—	—	—
openrouter/pareto-code Pareto Code Router	2M	$-1000000 / 1M	—	—	—
Perceptron · 1 model
perceptron/perceptron-mk1 Perceptron: Perceptron Mk1	33K	$0.15 / 1M	—	—	—
Prime Intellect · 1 model
prime-intellect/intellect-3 Prime Intellect: INTELLECT-3	131K	$0.2 / 1M	$0.2 / 1Mq8 · Nebius	$0.2 / 1M	soon
Reka · 2 models
rekaai/reka-edge Reka Edge	16K	$0.1 / 1M	—	—	—
rekaai/reka-flash-3 Reka Flash 3	66K	$0.1 / 1M	$0.1 / 1Mq8 · Reka	$0.1 / 1M	soon
Relace · 2 models
relace/relace-apply-3 Relace: Relace Apply 3	256K	$0.85 / 1M	$0.85 / 1Mq8 · Relace	$0.85 / 1M	—
relace/relace-search Relace: Relace Search	256K	$1.00 / 1M	—	—	—
Sao10k · 5 models
sao10k/l3-lunaris-8b Sao10K: Llama 3 8B Lunaris	8K	$0.04 / 1M	$0.04 / 1Mq8 · DeepInfra	$0.04 / 1M	soon
sao10k/l3-euryale-70b Sao10k: Llama 3 Euryale 70B v2.1	8K	$1.48 / 1M	—	—	—
sao10k/l3.1-70b-hanami-x1 Sao10K: Llama 3.1 70B Hanami x1	16K	$3.00 / 1M	—	—	—
sao10k/l3.1-euryale-70b Sao10K: Llama 3.1 Euryale 70B v2.2	131K	$0.85 / 1M	$0.85 / 1Mq8 · DeepInfra	$0.85 / 1M	soon
sao10k/l3.3-euryale-70b Sao10K: Llama 3.3 Euryale 70B	131K	$0.65 / 1M	—	—	—
StepFun · 1 model
stepfun/step-3.5-flash StepFun: Step 3.5 Flash	262K	$0.09 / 1M	$0.09 / 1Mq8 · DeepInfra	$0.09 / 1M	soon
Switchpoint · 1 model
switchpoint/router Switchpoint Router	131K	$0.85 / 1M	—	—	—
Tencent · 2 models
tencent/hunyuan-a13b-instruct Tencent: Hunyuan A13B Instruct	131K	$0.14 / 1M	$0.14 / 1Mq8 · SiliconFlow	$0.14 / 1M	soon
tencent/hy3-preview Tencent: Hy3 preview	262K	$0.066 / 1M	—	—	—
TheDrummer · 4 models
thedrummer/cydonia-24b-v4.1 TheDrummer: Cydonia 24B V4.1	131K	$0.3 / 1M	—	—	—
thedrummer/rocinante-12b TheDrummer: Rocinante 12B	33K	$0.17 / 1M	—	—	—
thedrummer/skyfall-36b-v2 TheDrummer: Skyfall 36B V2	33K	$0.55 / 1M	$0.55 / 1Mq8 · Parasail	$0.55 / 1M	soon
thedrummer/unslopnemo-12b TheDrummer: UnslopNemo 12B	33K	$0.4 / 1M	$0.4 / 1Mq8 · NextBit	$0.4 / 1M	soon
Undi95 · 1 model
undi95/remm-slerp-l2-13b ReMM SLERP 13B	6K	$0.45 / 1M	$0.5 / 1Mq8 · Mancer 2	$0.5 / 1M	soon
Upstage · 1 model
upstage/solar-pro-3 Upstage: Solar Pro 3	128K	$0.15 / 1M	—	—	—
Writer · 1 model
writer/palmyra-x5 Writer: Palmyra X5	1.0M	$0.6 / 1M	—	—	—
Xiaomi · 5 models
xiaomi/mimo-v2-flash Xiaomi: MiMo-V2-Flash	262K	$0.1 / 1M	$0.1 / 1Mq8 · Xiaomi	$0.1 / 1M	—
xiaomi/mimo-v2-omni Xiaomi: MiMo-V2-Omni	262K	$0.4 / 1M	$0.4 / 1Mq8 · Xiaomi	$0.4 / 1M	—
xiaomi/mimo-v2-pro Xiaomi: MiMo-V2-Pro	1.0M	$1.00 / 1M	$1.00 / 1Mq8 · Xiaomi	$1.00 / 1M	—
xiaomi/mimo-v2.5 Xiaomi: MiMo-V2.5	1.0M	$0.14 / 1M	$0.14 / 1Mq8 · Xiaomi	$0.14 / 1M	—
xiaomi/mimo-v2.5-pro Xiaomi: MiMo-V2.5-Pro	1.0M	$0.435 / 1M	$0.435 / 1Mq8 · Xiaomi	$0.435 / 1M	—
Z.AI · 12 models
z-ai/glm-4-32b Z.ai: GLM 4 32B	128K	$0.1 / 1M	—	—	—
z-ai/glm-4.5 Z.ai: GLM 4.5	131K	$0.6 / 1M	$0.6 / 1Mq8 · Novita	$0.6 / 1M	soon
z-ai/glm-4.5-air Z.ai: GLM 4.5 Air	131K	$0.125 / 1M	$0.125 / 1Mq8 · Io Net	$0.125 / 1M	soon
z-ai/glm-4.5v Z.ai: GLM 4.5V	66K	$0.6 / 1M	$0.6 / 1Mq8 · Novita	$0.6 / 1M	soon
z-ai/glm-4.6 Z.ai: GLM 4.6	203K	$0.43 / 1M	$0.43 / 1Mq4 · DeepInfra	$0.43 / 1M	soon
z-ai/glm-4.6v Z.ai: GLM 4.6V	131K	$0.3 / 1M	$0.3 / 1Mq8 · Z.AI	$0.3 / 1M	soon
z-ai/glm-4.7 Z.ai: GLM 4.7	203K	$0.4 / 1M	$0.4 / 1Mq4 · DeepInfra	$0.4 / 1M	soon
z-ai/glm-4.7-flash Z.ai: GLM 4.7 Flash	203K	$0.06 / 1M	$0.125 / 1Mq8 · Venice	$0.125 / 1M	soon
z-ai/glm-5 Z.ai: GLM 5	203K	$0.6 / 1M	$0.6 / 1Mq4 · DeepInfra	$0.6 / 1M	soon
z-ai/glm-5-turbo Z.ai: GLM 5 Turbo	203K	$1.20 / 1M	$1.20 / 1Mq8 · AtlasCloud	$1.20 / 1M	soon
z-ai/glm-5.1 Z.ai: GLM 5.1	203K	$0.98 / 1M	$0.98 / 1Mq8 · Baidu	$0.98 / 1M	soon
z-ai/glm-5v-turbo Z.ai: GLM 5V Turbo	203K	$1.20 / 1M	$1.20 / 1Mq8 · Z.AI	$1.20 / 1M	soon

Prices are input tokens per 1M. Standard / Batch show the cheapest q8/q4 provider variant Atlas would route to today; the decision is made per call from your operation’s eval-gate history. Atlas Network rates publish once the partner desktop app ships out of invite-only beta.

On Pay as you go, what you see is what you pay (and often less, when Atlas routes to a cheaper variant). Reliability Loop adds a 10% markup itemised separately on every invoice in exchange for the per-operation tuning, eval-gated refund, and LoRA training. Card statement descriptor: NEWMEN.AI*CREDITS. Auto top-up keeps your balance above zero so calls never fail mid-flight; thresholds are configurable in /console/settings/billing.

Talk to a solutions engineer

Atlas is sold to teams who commit to meaningful production volume. That commitment unlocks the reliability loop.