Skip to main content
BenchLM
Data

Ornith-1.5-397B vs Qwen3.5-27B

Updated October 2, 2026. Rank says Ornith-1.5-397B is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.

Share or export
Share on XLinkedInSocial cardCSVAPI/MCP

Decision reading

Ornith-1.5-397B has the higher public point estimate, 62.72 versus 48.33. Their conditional score ranges overlap. These ranges do not establish rank confidence. 3 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.

Model A

Ornith AI

62.72/100

Estimated · Public rank #45

Conditional range 48.4–77.1

Model B
Alibaba logo

Alibaba

48.33/100

Estimated · Public rank #98

Conditional range 38.6–58.0

Shared results
3
Ornith-1.5-397B only
15
Qwen3.5-27B only
13
Like-for-like categories
0 / 8
Estimated: Ornith-1.5-397B and Qwen3.5-27B. Conditional ranges do not establish rank confidence.How the comparison works

Which one for your work

Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.

No workload recommendation clears the current evidence threshold.

Use the matched evidence, workload costs, and sourced specifications below instead of treating a point score as a universal answer.

Show secondary and unsupported calls
  • Coding work

    Code generation, repair, and software-engineering tasks

    Not enough matched evidence

    Ornith-1.5-397B and Qwen3.5-27B are scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.

    Confidence: limited
  • Agentic work

    Tool use, computer use, and multi-step task completion

    Not enough matched evidence

    Ornith-1.5-397B and Qwen3.5-27B are scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.

    Confidence: limited
  • Long documents

    Prompts that approach the documented context limit

    No clear pick

    The documented context windows are equal.

    Confidence: documented
  • Chat turn cost

    1K fresh input + 500 output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates
  • Cache-heavy agent loop cost

    200K cached + 20K fresh input + 10K output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates
  • Repository review cost

    50K fresh input + 3K output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates

Which one for a specific job

Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.

The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.

54.7Ornith-1.5-397B34.9Qwen3.5-27B

Directional only · BenchAlign v5.8

Ornith-1.5-397B has the higher coding point estimate. Conditional score ranges do not establish rank confidence.

Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.

Same basis rules as the category table below

What is actually comparable

Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.

3 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.

Shape of the matched evidence

Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.

Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.

Bars run 0–100 on each benchmark’s normalized display scale

Category results, on a stated basis

Each row shows the public-lane category score for both models: the BenchAlign v5.8 lane for agentic, coding, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.

Agentic

Directional only
Ornith-1.5-397B
56.8
Estimated · #29/119
Qwen3.5-27B
35.1
Estimated · #70/119
Basis
BenchAlign v5.8 lane · 7 vs 4 public rows
Reading
Directional only

Coding

Directional only
Ornith-1.5-397B
54.7
Estimated · #34/144
Qwen3.5-27B
34.9
Estimated · #83/144
Basis
BenchAlign v5.8 lane · 7 vs 2 public rows
Reading
Directional only

Knowledge

Directional only
Ornith-1.5-397B
63.3
Estimated · #36/171
Qwen3.5-27B
44.4
Supported · #83/171
Basis
BenchAlign v5.8 lane · 4 vs 3 public rows
Reading
Directional only

Reasoning

Not comparable
Ornith-1.5-397B
Not ranked
Qwen3.5-27B
52.7
Unranked · 3 rankable rows
Basis
Provisional lane · 0 vs 1 weighted rows
Reading
Not comparable

Multimodal

Not comparable
Ornith-1.5-397B
Not ranked
Qwen3.5-27B
71.6
Unranked · 3 rankable rows
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Multilingual

Not comparable
Ornith-1.5-397B
Not ranked
Qwen3.5-27B
87.1
#10/16
Basis
Provisional lane · 0 vs 1 weighted rows
Reading
Not comparable

Instruction following

Not comparable
Ornith-1.5-397B
Not ranked
Qwen3.5-27B
91.5
#12/125
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Math

Not comparable
Ornith-1.5-397B
Not ranked
Qwen3.5-27B
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign v5.8) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.

Supported evidence per lane · bars run 0–100Methodology

What each workload costs

Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.

Chat turn

1K fresh input + 500 output tokens

Ornith-1.5-397B
Self-hosted; infrastructure cost varies
Fits in one request
Qwen3.5-27B
Self-hosted; infrastructure cost varies
Fits in one request

Ornith-1.5-397B has no comparable published API token rate. Qwen3.5-27B has no comparable published API token rate.

Repository review

50K fresh input + 3K output tokens

Ornith-1.5-397B
Self-hosted; infrastructure cost varies
Fits in one request
Qwen3.5-27B
Self-hosted; infrastructure cost varies
Fits in one request

Ornith-1.5-397B has no comparable published API token rate. Qwen3.5-27B has no comparable published API token rate.

Cache-heavy agent loop

200K cached + 20K fresh input + 10K output tokens

Ornith-1.5-397B
Self-hosted; infrastructure cost varies
Fits in one request
Cached-input rate unavailable
Qwen3.5-27B
Self-hosted; infrastructure cost varies
Fits in one request
Cached-input rate unavailable

Ornith-1.5-397B has no comparable published API token rate. Qwen3.5-27B has no comparable published API token rate.

Cached input falls back to the list input rate only where a cached rate is unpublished

Specification differences

Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.

Context window

Maximum documented context; output-token limits may be lower.

Ornith-1.5-397B

Qwen3.5-27B

262K

API model ID

Ornith-1.5-397B

Not sourced

Qwen3.5-27B

Not sourced

Cached-input rate

A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.

Ornith-1.5-397B

No comparable hosted API rate

Ornith-1.5-397B model card

Qwen3.5-27B

No comparable hosted API rate

Documented inputs

Ornith-1.5-397B

Not sourced

Qwen3.5-27B

Not sourced

Documented outputs

Ornith-1.5-397B

Not sourced

Qwen3.5-27B

Not sourced

Provider availability

Ornith-1.5-397B

Not sourced

Qwen3.5-27B

Not sourced

Reasoning profile

Ornith-1.5-397B

Reasoning

Qwen3.5-27B

Reasoning

Weight access

Ornith-1.5-397B

Open Weight

Qwen3.5-27B

Open Weight

License

Ornith-1.5-397B

Open Weight

Qwen3.5-27B

Open Weight

Release date

Ornith-1.5-397B

2026-08-18

Qwen3.5-27B

2026-03-04

If you already use one of these models

Deployment change
The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
Quality signal
Ornith-1.5-397B has the higher public point estimate, 62.72 versus 48.33. Their conditional score ranges overlap. These ranges do not establish rank confidence.
Workload cost
A complete comparable API-rate estimate is not available for both models.
Context tradeoff
Both models list 262K.
Run the same representative tasks against both endpoints before changing production traffic.

Questions

Which is better, Ornith-1.5-397B or Qwen3.5-27B?

Ornith-1.5-397B has the higher public point estimate, 62.72 versus 48.33. Their conditional score ranges overlap. These ranges do not establish rank confidence. The page therefore keeps the decision tied to the specific documented workload.

Which is better for coding, Ornith-1.5-397B or Qwen3.5-27B?

Ornith-1.5-397B scores higher for coding on the public lane, 54.7 to 34.9. Ornith-1.5-397B and Qwen3.5-27B are scored on Estimated evidence for coding, so the reading is directional rather than like-for-like. BenchLM does not name a winner for coding from a directional reading; read the shared benchmark rows directly and test the models on the same task set.

Which is better for agentic tasks, Ornith-1.5-397B or Qwen3.5-27B?

Ornith-1.5-397B scores higher for agentic tasks on the public lane, 56.8 to 35.1. Ornith-1.5-397B and Qwen3.5-27B are scored on Estimated evidence for agentic tasks, so the reading is directional rather than like-for-like. BenchLM does not name a winner for agentic tasks from a directional reading; read the shared benchmark rows directly and test the models on the same task set.

Which costs less, Ornith-1.5-397B or Qwen3.5-27B?

Both models do not have comparable published API token rates, so this page does not name a universal price winner.

Which has the larger context window, Ornith-1.5-397B or Qwen3.5-27B?

Both models list the same context window, 262K.

Benchmark evidence

The full public result ledger is available for audit without forcing a wide desktop table onto a phone.

Browse raw public benchmark evidence31 rows

Agentic

  • Terminal-Bench 2.1

    Ornith-1.5-397B86.1%
    Source
    Qwen3.5-27B—

    Not directly comparable

  • HLE w/ tools

    Ornith-1.5-397B56.1%
    Source
    Qwen3.5-27B—

    Not directly comparable

  • MCP Atlas

    Ornith-1.5-397B80%
    Source
    Qwen3.5-27B—

    Not directly comparable

  • Toolathlon-Verified

    Ornith-1.5-397B71.2%
    Source
    Qwen3.5-27B—

    Not directly comparable

  • WideResearch

    Ornith-1.5-397B80.8%
    Source
    Qwen3.5-27B—

    Not directly comparable

  • BrowseComp

    Ornith-1.5-397B86.6%
    Source
    Qwen3.5-27B61%
    Source

    Ornith-1.5-397B leads this result

  • Claw-Eval

    Ornith-1.5-397B81.4%
    Source
    Qwen3.5-27B—

    Not directly comparable

  • Terminal-Bench 2.0

    Ornith-1.5-397B—
    Qwen3.5-27B41.6%
    Source

    Not directly comparable

  • OSWorld-Verified

    Ornith-1.5-397B—
    Qwen3.5-27B56.2%
    Source

    Not directly comparable

  • Gert Labs

    Ornith-1.5-397B—
    Qwen3.5-27B39.41%
    Source

    Not directly comparable

Coding

  • Terminal-Bench 2.1

    Ornith-1.5-397B86.1%
    Source
    Qwen3.5-27B—

    Not directly comparable

  • SWE-bench Verified

    Ornith-1.5-397B86%
    Source
    Qwen3.5-27B72.4%
    Source

    Ornith-1.5-397B leads this result

  • SWE-bench Pro

    Ornith-1.5-397B65.1%
    Source
    Qwen3.5-27B—

    Not directly comparable

  • SWE Multilingual

    Ornith-1.5-397B79.6%
    Source
    Qwen3.5-27B—

    Not directly comparable

  • DeepSWE

    Ornith-1.5-397B56.0%
    Source
    Qwen3.5-27B—

    Not directly comparable

  • frontierBench

    Ornith-1.5-397B13.5%
    Source
    Qwen3.5-27B—

    Not directly comparable

  • NL2Repo

    Ornith-1.5-397B59.5%
    Source
    Qwen3.5-27B—

    Not directly comparable

  • SWE-Rebench

    Ornith-1.5-397B—
    Qwen3.5-27B58.9%
    Source

    Not directly comparable

Reasoning

  • LongBench v2

    Ornith-1.5-397B—
    Qwen3.5-27B60.6%
    Source

    Not directly comparable

Multimodal

  • MMMU

    Ornith-1.5-397B—
    Qwen3.5-27B82.3%
    Source

    Not directly comparable

  • MMVU

    Ornith-1.5-397B—
    Qwen3.5-27B73.3%
    Source

    Not directly comparable

  • MathVision

    Ornith-1.5-397B—
    Qwen3.5-27B86.0%
    Source

    Not directly comparable

  • V*

    Ornith-1.5-397B—
    Qwen3.5-27B93.7%
    Source

    Not directly comparable

Knowledge

  • GPQA

    Ornith-1.5-397B92.8%
    Source
    Qwen3.5-27B85.5%
    Source

    Ornith-1.5-397B leads this result

  • GPQA-D

    Ornith-1.5-397B92.8%
    Source
    Qwen3.5-27B—

    Not directly comparable

  • HLE

    Ornith-1.5-397B44.6%
    Source
    Qwen3.5-27B—

    Not directly comparable

  • HLE w/o tools

    Ornith-1.5-397B44.6%
    Source
    Qwen3.5-27B—

    Not directly comparable

  • MMLU-Pro

    Ornith-1.5-397B—
    Qwen3.5-27B86.1%
    Source

    Not directly comparable

  • SuperGPQA

    Ornith-1.5-397B—
    Qwen3.5-27B65.6%
    Source

    Not directly comparable

Multilingual

  • MMLU-ProX

    Ornith-1.5-397B—
    Qwen3.5-27B82.2%
    Source

    Not directly comparable

Instruction following

  • IFEval

    Ornith-1.5-397B—
    Qwen3.5-27B95%
    Source

    Not directly comparable

31 public results · 3 shared

Watch Ornith-1.5-397B vs Qwen3.5-27B

One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.

Read a sample issue

Join 2,000+ readers.

Last updated October 2, 2026