Data as of October 20, 2026 · How the score is built
Ornith-1.5-122B
Decision readingOrnith-1.5-122B is tracked, but not publicly ranked yet. The original benchmark tables below include the published metrics for this evaluated configuration. Their distinct protocols remain outside general model rankings.
Original benchmark results
Every published metric for this configuration, with source precision and evaluation limits. Download results (JSON) · Numeric metrics (CSV)
Ornith-reported benchmark results
Ornith reports these results for its 122B checkpoint. They remain outside general rankings. The original source does not label a common unit or publish evaluation dates for these rows.
Printed precision is preserved. Terminal-Bench uses Claude Code; the SWE-bench rows apply the source's anti-hacking rules. SciCode's track is unreported. WideSearch remains distinct from WideResearch. Context extensions used in an evaluation do not change the checkpoint's native 262,144-token context limit.
| Benchmark | Published value | Evaluation setup |
|---|---|---|
| TerminalBench 2.1 | 81.200 | Claude Code 2.1.126; Parser: json; Temperature: 1; Top-p: 1; Output token limit: 131072; Runs: 5 |
| CyberGym | 80.000 | Not reported |
| NL2Repo | 51.200 | Temperature: 1; Top-p: 1; Evaluation context: 400K; Evaluation output: 48K; Specified repositories and pip packages blocked: yes |
| SWE-bench Pro (no-hack) | 76.900 | OpenHands; Temperature: 1; Top-p: 0.95; Evaluation context: 256K; Anti-hacking safeguards: yes |
| SWE-bench Verified (non-hack) | 84.500 | OpenHands; Temperature: 1; Top-p: 0.95; Evaluation context: 256K; Anti-hacking safeguards: yes |
| SWE-bench Multilingual (non-hack) | 76.400 | OpenHands; Temperature: 1; Top-p: 0.95; Evaluation context: 256K; Anti-hacking safeguards: yes |
| HLE (with tools) | 48.000 | Tools: yes; Judge: Claude 4.6 Opus |
| HLE (w/o tools) | 37.680 | Tools: no; Judge: Claude 4.6 Opus |
| SciCode | 53.800 | Not reported |
| WideSearch | 75.200 | Not reported |
| BrowseComp | 78.500 | Not reported |
| MCP-Atlas | 74.600 | Thinking: yes; Public tasks: 500; Task timeout (minutes): 10; Judge: Claude 4.8 Opus |
| Toolathlon-Verified | 70.800 | Official evaluation service; Evaluation token limit: 128K |
| ClawEval | 79.400 | Temperature: 0.6; Evaluation context: 256K |
Lineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- Weights model ID
- ornith-ai/Ornith-1.5-122BOrnith official model repository
- Context window
- 262144Ornith model configuration: 262,144 tokens
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- text, image, videoOrnith model configuration
- Output modalities
- textOrnith model configuration
- Parameters
- 125.1B supplied parameters; MoE active count unreportedHugging Face safetensors metadata: 125,086,497,008 BF16 parameters
- Availability
- Hugging FaceOrnith model card
- Cloud regions
- Not tracked yet
- Lifecycle
- Tracked
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Not documented in the pricing recordOrnith-1.5-122B official model card
- Self-host
- Open weights available; hardware estimate not sourced
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
The original benchmark tables show this evaluated configuration’s published results. Training choices, prompts, response splits, and metrics remain attached to each table. The profile stays outside general model rankings.
Ornith-1.5-122B is a open weight model with a 262144 context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
The model card declares MIT-licensed open weights, but its linked LICENSE file is unavailable at the reviewed revision. The supplied BF16 checkpoint contains about 125.1 billion parameters. A first-party hosted API token price and active parameter count are unreported.
The original-results table retains all 14 Ornith-reported values and their disclosed evaluation settings. Claude Code, OpenHands anti-hacking variants, and unspecified SciCode settings remain explicit; no result is transferred into weighted benchmark slots. Native context is 262,144 tokens; longer contexts require optional scaling. The exact base checkpoint and public release date are unconfirmed.
Ornith-1.5-122B sits in the Ornith 1.5 family with Ornith-1.5-397B, Ornith-1.5-35B-A3B, Ornith-1.5-9B. The original benchmark results appear in the source tables above; they remain separate from weighted benchmark slots.
Last updated October 20, 2026. Runtime fields remain blank until a sourced snapshot exists.
Deployment options
Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.
Published weights are available, but BenchLM does not yet have a sourced parameter and VRAM profile for this exact model. Hardware cost estimates stay unavailable until that sizing record is complete.
Estimate VRAM from known parametersQuestions
How does Ornith-1.5-122B perform overall in AI benchmarks?
This configuration has original benchmark results in the source tables on this profile. Every published metric retains its evaluated configuration, source precision, and protocol. These results stay outside general model rankings; they do not produce a public overall score or transfer to a base model with different settings.
Is Ornith-1.5-122B open source?
Ornith-1.5-122B is an open-weight model from Ornith AI. Its weights can be downloaded for local or hosted deployment, subject to the published license. Open weight does not automatically mean open source: training data and training code may remain private, and commercial restrictions can still apply.
Which sibling models are related to Ornith-1.5-122B?
Ornith-1.5-122B belongs to the Ornith 1.5 family. Related tracked variants include Ornith-1.5-397B, Ornith-1.5-35B-A3B, Ornith-1.5-9B. A sibling link indicates shared lineage or a documented configuration relationship; it does not mean the variants have identical pricing, context limits, benchmark evidence, or deployment behavior. Compare before switching.
Does Ornith-1.5-122B have full benchmark coverage on BenchLM?
This configuration has results in the original benchmark tables on this profile. Those tables preserve every imported source metric, but they do not fill the site’s general scoring slots. Other benchmarks remain unmeasured, and compatible configurations are required before comparing results across different reports.
What is the context window size of Ornith-1.5-122B?
Ornith-1.5-122B has a documented context window of 262144. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.