Muse Glimmer 30B vs Ornith 1.5 35B A3B vs Toast 1
Too close to call on our weighted score (Ornith 1.5 35B A3B 59, Muse Glimmer 30B 57, Toast 1 47). The right pick depends on what you value most.
MetaMuse Glimmer 30B
57/100- ECI—
- Price$0.30 / $1.20
- Context131K
DeepReinforce
Ornith 1.5 35B A3B
59/100- ECI—
- Price$0.20 / $1.70
- Context262K
Mixedbread
Toast 1
47/100- ECI—
- Price$0.30 / $0.72
- Context131K
Too close to call
It is close. Our weighted score puts them within 1 points (Ornith 1.5 35B A3B 59/100, Muse Glimmer 30B 57/100, Toast 1 47/100), so choose by what matters most for your work: Toast 1 on price and Ornith 1.5 35B A3B for long inputs. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
- CapabilityNot enough dataNo independent benchmark covers every model here yet
- Lowest priceToast 1Toast 1 $0.405 · Muse Glimmer 30B $0.525 · Ornith 1.5 35B A3B $0.575 per 1M tokens (3:1 blend)
- Longest contextOrnith 1.5 35B A3BOrnith 1.5 35B A3B 262,144 · Muse Glimmer 30B 131,072 · Toast 1 131,000 tokens
- Widest inputsMuse Glimmer 30B and Ornith 1.5 35B A3BMuse Glimmer 30B: Text, Images · Ornith 1.5 35B A3B: Text, Images · Toast 1: Text
- Self-hostingMuse Glimmer 30B and Ornith 1.5 35B A3BPublishes downloadable weights (Apache 2.0 and MIT)
| Measure | Weight | Muse Glimmer 30B | Ornith 1.5 35B A3B | Toast 1 |
|---|---|---|---|---|
| Price | 50% | 63 | 61 | 68 |
| Inputs & features | 30% | 70 | 70 | 25 |
| Context window | 20% | 24 | 37 | 24 |
| Overall | 100% | 57/100 | 59/100 | 47/100 |
Left out because at least one model lacks the data: capability. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | Muse Glimmer 30BMeta | Ornith 1.5 35B A3BDeepReinforce | Toast 1Mixedbread |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $0.30 | $0.20 (best) | $0.30 |
| Output | $1.20 | $1.70 | $0.72 (best) |
| Cached input | — | — | — |
| Blended (3:1) | $0.525 | $0.575 | $0.405 (best) |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 10 providers | Median of 2 providers | Median of 1 providers |
| Limits | |||
| Context window | 131,072 tokens | 262,144 tokens (best) | 131,000 tokens |
| Max output | 131,072 tokens (best) | — | 4,000 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | Yes | No |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | Yes | Yes | No |
| Tool calling | Yes | Yes | Yes |
| Structured output | Yes | Yes | No |
| Availability | |||
| Weights | OpenApache 2.0 | OpenMIT | Proprietary |
| API model ID | — | — | — |
| API providers | 12 (best) | 2 | 1 |
| Released | Aug 10, 2026 | Aug 18, 2026 | Aug 13, 2026 |
| Knowledge cutoff | Jan 4, 2026 | — | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Muse Glimmer 30B$5.40
- Ornith 1.5 35B A3B$5.40
- Toast 1$4.44
Which should you choose?
Which is better: Muse Glimmer 30B, Ornith 1.5 35B A3B or Toast 1?
It is close. Our weighted score puts them within 1 points (Ornith 1.5 35B A3B 59/100, Muse Glimmer 30B 57/100, Toast 1 47/100), so choose by what matters most for your work: Toast 1 on price and Ornith 1.5 35B A3B for long inputs. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, Muse Glimmer 30B, Ornith 1.5 35B A3B or Toast 1?
Toast 1 is cheaper at $0.30 input / $0.72 output per million tokens (median across 1 API provider). Muse Glimmer 30B costs $0.30 input / $1.20 output per million tokens (median across 10 API providers); Ornith 1.5 35B A3B costs $0.20 input / $1.70 output per million tokens (median across 2 API providers). At a typical mix of three input tokens to one output token, that is $0.405 per million tokens for Toast 1 versus $0.525 for Muse Glimmer 30B (1.3× as much) and $0.575 for Ornith 1.5 35B A3B (1.4× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Muse Glimmer 30B has not been scored yet, Ornith 1.5 35B A3B has not been scored yet and Toast 1 has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Muse Glimmer 30B, Ornith 1.5 35B A3B and Toast 1 yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. All three support tool calling for agent workflows.
Which has the bigger context window?
Ornith 1.5 35B A3B has the largest context window at 262,144 tokens, against 131,072 for Muse Glimmer 30B and 131,000 for Toast 1. Maximum output per response: Muse Glimmer 30B up to 131,072, Toast 1 up to 4,000 tokens.
Which can read images, PDFs, audio or video?
Muse Glimmer 30B accepts text and images; Ornith 1.5 35B A3B accepts text and images; Toast 1 accepts text. Muse Glimmer 30B handles the widest range of inputs.
Are any of these open source?
Muse Glimmer 30B and Ornith 1.5 35B A3B publishes its weights (Apache 2.0 and MIT) and can be self-hosted; Toast 1 is proprietary.
Which is newer?
Ornith 1.5 35B A3B is the newest, released Aug 18, 2026. Toast 1 came out Aug 13, 2026; Muse Glimmer 30B came out Aug 10, 2026. Knowledge cutoff: Muse Glimmer 30B Jan 4, 2026.
How do you decide the winner?
Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Oct 4, 2026.
