Models for octomind.
Zero keys.
Skip collecting API keys from five providers. One login — every model works, in the CLI, in your editor and in any OpenAI-compatible client, pay as you go from credits. Star muvon/octomind on GitHub for a one-time $5 starter credit. Open to everyone today — no invite, no waitlist.
$ curl -fsSL https://octomind.run/install.sh | bash
$ octomind login # approve in the browser — no keys to paste
$ octomind # models just work — pick any with -m octohub:…
Transparent prices
Published per-token prices, metering that matches the table exactly. Caps you can see live — never a surprise bill.
$5 to start
Star muvon/octomind on GitHub and get a one-time $5 starter credit. Every model at the provider's price plus 5% — no card, no expiry.
One key for models everywhere
The same hub key serves model calls in the CLI, inside cloud machines, and from any OpenAI-compatible tool pointed at the endpoint. (Scripting machines and sessions is the separate Developer API, documented in the panel.)
The table
35 models · 10 embedding models. Every model on every account, at the provider's price plus 5%, pay as you go from credits. Same table the panel and the meter use — snapshot 2026-10-12.
The octomind benchmark ran on glm-5.2, which solved 24/25 real-PR benchmark tasks running in octomind, ahead of Claude Code with Opus.
Swipe the table for output prices and context →
| Model | Input /1M | Output /1M | Context |
|---|---|---|---|
| DeepSeek V4 Flash | $0.44 | $1.32 | 1M |
| DeepSeek V4 Pro | $1.32 | $3.96 | 1M |
| DeepSeek V4.1 Flash | $0.30 | $1.20 | 1M |
| Qwen3.7 Plus | $0.40 | $1.60 | 256k |
| Qwen3.8 Omni Flash | $0.15 | $0.47 | 1M |
| Qwen3.5 397B | $0.60 | $3.60 | 256k |
| Gemma 4 31B | $0.20 | $0.20 | 128k |
| GPT-OSS 20B | $0.03 | $0.10 | 128k |
| GPT-OSS 120B | $0.35 | $0.75 | 128k |
| MiniMax M3 | $0.30 | $1.20 | 1M |
| Kimi 2.6 | $0.95 | $4.00 | 256k |
| Kimi 2.7 Code | $0.95 | $4.00 | 256k |
| GLM-5.3 | $1.40 | $4.40 | 1M |
| GLM-5.3 Flash | $0.15 | $0.50 | 1M |
| Mistral Large 4 | $1.36 | $4.18 | 1M |
| Inkling | $1.00 | $4.05 | 1M |
| Inkling Small | $0.50 | $1.20 | 1M |
| Mercury 2.5 | $0.04 | $0.15 | 260k |
| Kimi K3 | $3.00 | $15.00 | 1M |
| Qwen3.8 Max | $2.00 | $6.00 | 1M |
| Muse Spark 1.3 | $1.25 | $4.25 | 1M |
| Muse Spark 1.3 Contributor | $0.10 | $0.20 | 1M |
| Grok 4.7 | $2.00 | $6.00 | 500k |
| Grok Build 0.1 | $1.00 | $2.00 | 256k |
| Claude Sonnet 5.5 | $2.00 | $10.00 | 1M |
| Claude Haiku 5.5 | $0.10 | $0.50 | 1M |
| Claude Opus 5.5 | $4.00 | $20.00 | 1M |
| Claude Fable 5.1 | $10.00 | $50.00 | 1M |
| GPT-6 Astra | $10.00 | $50.00 | 1M |
| GPT-6.1 Sol | $2.00 | $10.00 | 1M |
| GPT-6 Luna | $0.10 | $0.50 | 1M |
| GPT 5.6 Terra | $2.00 | $12.00 | 1M |
| Gemini 3.8 Flash | $0.75 | $3.75 | 1M |
| Gemini 3.5 Flash Lite | $0.30 | $2.50 | 1M |
| Gemini 3.1 Pro | $2.00 | $12.00 | 1M |
| Voyage 4 Large | $0.12 | — | 32k |
| Voyage 4 | $0.06 | — | 32k |
| Voyage 4 Lite | $0.02 | — | 32k |
| Voyage Code 4 | $0.12 | — | 32k |
| Jina v5 Omni Small | $0.05 | — | 32k |
| Jina v5 Omni Nano | $0.02 | — | 8k |
| Jina v5 Text Small | $0.05 | — | 32k |
| Jina v5 Text Nano | $0.02 | — | 8k |
| OpenAI Embedding 3 Small | $0.02 | — | 8k |
| OpenAI Embedding 3 Large | $0.13 | — | 8k |
Prices are what you pay: the provider's price plus 5%, drawn from plan usage and then credits. The hub is open to every account; top up credits from $5. Prices track the octomind version we run.
Embeddings route through the hub too — Voyage, Jina and OpenAI models from $0.02/1M tokens, drawn from plan usage and then credits. Local CPU embeddings stay free on every plan, hub or not.
Fair use is per account, not per key: 1 model call at a time on the $5 starter, 2 after any top-up, up to 5 as lifetime top-ups grow — 3 in parallel on Pro, 15 pooled on Team — past that, requests queue briefly and then 429.
Do the maths with real numbers
Same per-token prices as the table — pick a model, size the session.
One endpoint, any tool
The hub is an LLM API gateway that speaks OpenAI-compatible HTTP. Point any client at it — your editor, your scripts, your own product — with the hub key from the panel's Keys page. Same key, same wallet, same live caps.
export OPENAI_BASE_URL=https://hub.octomind.run/v1
export OPENAI_API_KEY=<hub key — panel → Keys>
Prefer to self-host?
The LLM gateway is octohub — open source, Rust, one binary. Run your own with your own provider keys; the hosted hub is the same code with the keys problem solved.
Questions, answered straight
Can I use the hub outside octomind?
Yes. The endpoint is OpenAI-compatible — point any client at https://hub.octomind.run/v1 with the hub key from the panel’s Keys page. Your editor, your scripts, your own product: same key, same wallet, same caps.
What happens when I hit a cap?
Usage is drawn from credits — top up $5 or more and keys never pause mid-session. If credits run out, keys pause until you top up; machines, files and session history stay intact.
Which models can I use?
Every model, on every account — billed at the provider’s price plus 5% from credits. Star muvon/octomind on GitHub for a one-time $5 starter credit; login is required (it’s the abuse gate); no card, no expiry.
How many requests can run at once?
Fair use is per account: 1 model call at a time on the $5 starter, 2 after any top-up, up to 5 as lifetime top-ups grow — 3 in parallel on Pro, 15 pooled on Team. Past that, requests queue briefly and then return 429.
How current are these prices?
This table is snapshotted from the live catalog at every build — the date sits right under the table, and it is the same catalog the meter bills from. Prices track the octomind version we run.
What about embeddings?
They route through the hub too — Voyage, Jina and OpenAI models from $0.02/1M tokens, drawn from plan usage and then credits. Local CPU embeddings stay free on every plan, hub or not.
Models on credits. Machines on a plan.
Any account uses every model from its own tools, pay as you go from credits. Cloud machines come with a plan — $20 your first month, then $50/mo — and on a paid plan one allowance covers both the models and the machines.