OptimToken › Models by maker › Meta
Meta (Muse) API pricing
As of 2026-11-21, the OptimToken catalogue holds 16 Meta models. By price tier, 1 is Premium, 4 Mid-range and 11 Budget. The cheapest is Llama 3.1 8B Instruct at $0.02 per 1M input tokens and $0.04 per 1M output tokens. The most expensive is Muse Wonder at $4.00 input and $20.00 output, 447x the cheapest on blended price (30% input, 70% output). These are the prices we quote: a model sold by several hosts has other rates, listed on its own page.
Meta models and prices
| Model | Input | Output | Batch (in / out) | Cache read | Context | Licence |
|---|---|---|---|---|---|---|
| Llama 3.1 8B Instruct | $0.02 | $0.04 | not published | not published | 131K | Open weights (Llama 3.x) |
| Muse Spark 1.2 Contributor | $0.10 | $0.20 | not published | $0.002 | 1M | Unknown |
| Muse Spark 1.3 Contributor | $0.10 | $0.20 | not published | $0.002 | 1M | Unknown |
| Llama Guard 4 12B | $0.18 | $0.18 | not published | not published | 164K | Open weights (Llama 4) |
| Llama 4 Scout | $0.10 | $0.30 | not published | not published | 1M | Open weights (Llama 4) |
| Llama 3.2 3B Instruct | $0.05 | $0.33 | not published | not published | 131K | Open weights (Llama 3.x) |
| Muse Ember | $0.10 | $0.50 | not published | $0.004 | 1M | Unknown |
| Llama 3.3 70B Instruct | $0.71 | $0.71 | not published | $0.71 | 131K | Open weights (Llama 3.x) |
| Llama 3.1 70B Instruct | $0.72 | $0.72 | not published | not published | 131K | Open weights (Llama 3.x) |
| Llama 4 Maverick | $0.35 | $1.00 | not published | $0.17 | 524K | Open weights (Llama 4) |
| Muse Glimmer 30B | $0.35 | $1.50 | $0.24 / $0.96 | $0.04 | 131K | Open source (Apache 2.0) |
| Muse Spark 1.1 | $1.25 | $4.25 | not published | $0.15 | 1M | Unknown |
| Muse Spark 1.2 | $1.25 | $4.25 | not published | $0.15 | 1M | Unknown |
| Muse Spark 1.3 | $1.25 | $4.25 | not published | $0.15 | 1M | Unknown |
| Muse Spark Open | $1.25 | $4.25 | not published | $0.15 | 1M | Open weights (Muse Model License) |
| Muse Wonder | $4.00 | $20.00 | not published | $0.20 | 1M | Unknown |
Published discounts: a prompt-cache read rate for 11 of the 16 and a batch rate for 1 of the 16. The most recent by release date is Muse Ember (2026-11). Arena ELO scores (as of 2026-11-11) exist for 11 of the 16; the highest is Muse Wonder at 1503.
Licences
Of the 16, 1 is open source: its weights are published under the Apache 2.0 licence, OSI-approved with no field-of-use limit, so it can be self-hosted as well as bought by the token. Another 8 are open weights, which is not the same as open source: the weights are published, but under the Llama 3.x, Llama 4 and Muse Model License licences, which restrict some uses. Read them before self-hosting commercially. No licence is recorded for the other 7, so this page does not say whether they can be self-hosted. Licences are as recorded in this catalogue: check the maker's own terms before relying on one.
Price changes since tracking began
Our daily price archive began on 2026-08-15 and holds a record for all 16 models. The price we quote has changed for 7 of them, 46 times in all, 34 of those changes flagged as a change of quoted host rather than a repricing. The most recent was to Llama 3.2 3B Instruct on 2026-11-20: -1.5% on the blended price (30% input, 70% output). Since tracking began, 10 of them joined the catalogue, most recently Muse Ember on 2026-11-10. Last checked against the 2026-11-21 snapshot.
Each row is a model's most recent change. Its own page has the full history.
| Model | Date | Old price (in / out) | New price (in / out) | Blended change | Changes in all |
|---|---|---|---|---|---|
| Llama 3.2 3B Instruct | 2026-11-20 | $0.0509 / $0.335 | $0.05 / $0.33 | -1.5% | 6 |
| Muse Glimmer 30B | 2026-11-17 | $0.30 / $1.20 | $0.35 / $1.50 | +24.2% (quoted host changed, not a repricing) | 11 |
| Llama 3.3 70B Instruct | 2026-11-10 | $0.59 / $0.79 | $0.71 / $0.71 | -2.7% (quoted host changed, not a repricing) | 14 |
| Llama 3.1 8B Instruct | 2026-11-09 | $0.05 / $0.08 | $0.02 / $0.04 | -52.1% (quoted host changed, not a repricing) | 3 |
| Llama 4 Maverick | 2026-11-05 | $0.1875 / $0.6525 | $0.35 / $1.00 | +56.9% (quoted host changed, not a repricing) | 5 |
| Llama 3.1 70B Instruct | 2026-11-03 | $0.40 / $0.40 | $0.72 / $0.72 | +80.0% (quoted host changed, not a repricing) | 3 |
| Llama 4 Scout | 2026-10-17 | $0.18 / $0.59 | $0.10 / $0.30 | -48.6% (quoted host changed, not a repricing) | 4 |
Head-to-head comparisons
Models from Meta appear on 1 ready-made side-by-side page.
Other makers
- AionLabs API pricing
- Alibaba (Qwen) API pricing
- Amazon (Nova) API pricing
- Anthropic (Claude) API pricing
- ByteDance (Seed) API pricing
- Cohere (Command) API pricing
- DeepSeek API pricing
- Google (Gemini) API pricing
- inclusionAI (Ling) API pricing
- MiniMax API pricing
- Mistral API pricing
- Moonshot (Kimi) API pricing
- Nex AGI API pricing
- Nvidia (Nemotron) API pricing
- OpenAI (GPT) API pricing
- Perplexity (Sonar) API pricing
- Sakana (Fugu) API pricing
- Tencent (Hy) API pricing
- Thinking Machines (Inkling) API pricing
- xAI (Grok) API pricing
- Xiaomi (MiMo) API pricing
- Zhipu (GLM) API pricing
Models by maker: every model in the catalogue, including makers with fewer than 3.
Your real cost also depends on caching and batching, negotiated discounts, marketplace agreements, and the contracts you already have. OptimNow audits exactly that.
Get a FinOps review