Agentic
Directional only- GPT-5.6 Sol
- 92.0
- Ornith-1.5-35B-A3B
- 67.6
- Weighted basis
- 2 vs 1 rows
- Reading
- Directional only
GPT-5.6 Sol vs Ornith-1.5-35B-A3B
Updated August 19, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
GPT-5.6 Sol has the higher public score, 82.04 versus 49.27, and the 90% score intervals do not overlap.
5 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.
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.
Long documents
Prompts that approach the documented context limit
GPT-5.6 Sol
GPT-5.6 Sol has the larger documented context window.
Confidence: documented
Show secondary and unsupported calls
Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.
Confidence: limited
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
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
27 GPT-5.6 Sol only
5 Shared
13 Ornith-1.5-35B-A3B only
- Shared results
- 5
- GPT-5.6 Sol only
- 27
- Ornith-1.5-35B-A3B only
- 13
- Like-for-like categories
- 0 / 8
3 categories use different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Category results, on a stated basis
Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.
Coding
Directional only- GPT-5.6 Sol
- 64.6
- Ornith-1.5-35B-A3B
- 71.5
- Weighted basis
- 1 vs 2 rows
- Reading
- Directional only
Knowledge
Directional only- GPT-5.6 Sol
- 94.6
- Ornith-1.5-35B-A3B
- 34.2
- Weighted basis
- 1 vs 2 rows
- Reading
- Directional only
Reasoning
Not comparable- GPT-5.6 Sol
- 92.5
- Ornith-1.5-35B-A3B
- Not measured
- Weighted basis
- 1 vs 0 rows
- Reading
- Not comparable
Math
Not comparable- GPT-5.6 Sol
- 87.5
- Ornith-1.5-35B-A3B
- Not measured
- Weighted basis
- 2 vs 0 rows
- Reading
- Not comparable
Multilingual
Not comparable- GPT-5.6 Sol
- Not measured
- Ornith-1.5-35B-A3B
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Multimodal
Not comparable- GPT-5.6 Sol
- 83.0
- Ornith-1.5-35B-A3B
- Not measured
- Weighted basis
- 1 vs 0 rows
- Reading
- Not comparable
Instruction following
Not comparable- GPT-5.6 Sol
- Not measured
- Ornith-1.5-35B-A3B
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
| Category | GPT-5.6 Sol | Ornith-1.5-35B-A3B | Weighted basis | Reading |
|---|---|---|---|---|
| Agentic | 92.0 | 67.6 | Directional only2 vs 1 rows | Directional only |
| Coding | 64.6 | 71.5 | Directional only1 vs 2 rows | Directional only |
| Knowledge | 94.6 | 34.2 | Directional only1 vs 2 rows | Directional only |
| Reasoning | 92.5 | Not measured | Not comparable1 vs 0 rows | Not comparable |
| Math | 87.5 | Not measured | Not comparable2 vs 0 rows | Not comparable |
| Multilingual | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
| Multimodal | 83.0 | Not measured | Not comparable1 vs 0 rows | Not comparable |
| Instruction following | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
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.
BrowseComp
Agentic
GPT-5.6 Sol: 92.2%Ornith-1.5-35B-A3B: 67.6%Normalized gap 24.6GPT-5.6 Sol source Ornith-1.5-35B-A3B sourceGPQA
Knowledge
GPT-5.6 Sol: 94.6%Ornith-1.5-35B-A3B: 89.2%Normalized gap 5.4GPT-5.6 Sol source Ornith-1.5-35B-A3B sourceSWE-bench Pro
Coding
GPT-5.6 Sol: 64.6%Ornith-1.5-35B-A3B: 59.6%Normalized gap 5.0GPT-5.6 Sol source Ornith-1.5-35B-A3B source
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
- GPT-5.6 Sol
- $0.02
- Fits in one request
- Ornith-1.5-35B-A3B
- Self-hosted; infrastructure cost varies
- Fits in one request
Ornith-1.5-35B-A3B has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- GPT-5.6 Sol
- $0.34
- Fits in one request
- Ornith-1.5-35B-A3B
- Self-hosted; infrastructure cost varies
- Fits in one request
Ornith-1.5-35B-A3B has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- GPT-5.6 Sol
- $0.5
- Fits in one request
- Ornith-1.5-35B-A3B
- Self-hosted; infrastructure cost varies
- Fits in one request
- Cached-input rate unavailable
Ornith-1.5-35B-A3B has no comparable published API token rate.
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationGPT-5.6 SolOrnith-1.5-35B-A3B
Context window
Maximum documented context; output-token limits may be lower.
GPT-5.6 Sol
1.05M
OpenAI model catalogOrnith-1.5-35B-A3B
API model ID
GPT-5.6 Sol
gpt-5.6-sol
OpenAI model catalogOrnith-1.5-35B-A3B
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
GPT-5.6 Sol
$0.5 per 1M cached input tokens
OpenAI pricingOrnith-1.5-35B-A3B
No comparable hosted API rate
Ornith-1.5-35B-A3B model cardDocumented inputs
GPT-5.6 Sol
text, image
OpenAI model catalogOrnith-1.5-35B-A3B
Not sourced
Documented outputs
GPT-5.6 Sol
Ornith-1.5-35B-A3B
Not sourced
Provider availability
GPT-5.6 Sol
Generally Available · OpenAI Responses API
OpenAI model catalogOrnith-1.5-35B-A3B
Not sourced
Reasoning profile
GPT-5.6 Sol
Reasoning
Ornith-1.5-35B-A3B
Reasoning
Weight access
GPT-5.6 Sol
Proprietary
Ornith-1.5-35B-A3B
Open Weight
License
GPT-5.6 Sol
Proprietary
Ornith-1.5-35B-A3B
Open Weight
Release date
GPT-5.6 Sol
2026-07-09
Ornith-1.5-35B-A3B
2026-08-18
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
- GPT-5.6 Sol has the higher public score, 82.04 versus 49.27, and the 90% score intervals do not overlap.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- Context tradeoff
- GPT-5.6 Sol has the larger documented window (1.05M).
Run the same representative tasks against both endpoints before changing production traffic.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence45 rows
Agentic
Terminal-Bench 3.0
GPT-5.6 Sol34.6%SourceOrnith-1.5-35B-A3B—Not directly comparable
Terminal-Bench 2.0
GPT-5.6 Sol91.9%SourceOrnith-1.5-35B-A3B—Not directly comparable
BrowseComp
GPT-5.6 Sol92.2%SourceOrnith-1.5-35B-A3B67.6%SourceGPT-5.6 Sol leads this result
OSWorld 2.0
GPT-5.6 Sol62.6%SourceOrnith-1.5-35B-A3B—Not directly comparable
CyberGym
GPT-5.6 Sol84.5%SourceOrnith-1.5-35B-A3B—Not directly comparable
ExploitGym
GPT-5.6 Sol33.7%SourceOrnith-1.5-35B-A3B—Not directly comparable
Toolathlon
GPT-5.6 Sol58%SourceOrnith-1.5-35B-A3B—Not directly comparable
Terminal-Bench 2.1
GPT-5.6 Sol—Ornith-1.5-35B-A3B67.8%SourceNot directly comparable
HLE w/ tools
GPT-5.6 Sol—Ornith-1.5-35B-A3B33.4%SourceNot directly comparable
MCP Atlas
GPT-5.6 Sol—Ornith-1.5-35B-A3B70.2%SourceNot directly comparable
Toolathlon-Verified
GPT-5.6 Sol—Ornith-1.5-35B-A3B48.7%SourceNot directly comparable
WideResearch
GPT-5.6 Sol—Ornith-1.5-35B-A3B67.8%SourceNot directly comparable
Claw-Eval
GPT-5.6 Sol—Ornith-1.5-35B-A3B72.5%SourceNot directly comparable
Coding
SWE-bench Pro
GPT-5.6 Sol64.6%SourceOrnith-1.5-35B-A3B59.6%SourceGPT-5.6 Sol leads this result
Terminal-Bench 2.0
GPT-5.6 Sol91.9%SourceOrnith-1.5-35B-A3B—Not directly comparable
deepSwe
GPT-5.6 Sol72.7%SourceOrnith-1.5-35B-A3B22%SourceGPT-5.6 Sol leads this result
FrontierCode 1.1 Extended
GPT-5.6 Sol60.6%SourceOrnith-1.5-35B-A3B—Not directly comparable
cursorBench32
GPT-5.6 Sol67.2%SourceOrnith-1.5-35B-A3B—Not directly comparable
VulcanBench v3
GPT-5.6 Sol87.0%SourceOrnith-1.5-35B-A3B—Not directly comparable
APEX-SWE
GPT-5.6 Sol45.8%SourceOrnith-1.5-35B-A3B—Not directly comparable
EEBench
GPT-5.6 Sol39.4%SourceOrnith-1.5-35B-A3B—Not directly comparable
CADGenBench Generation
GPT-5.6 Sol37.1%SourceOrnith-1.5-35B-A3B—Not directly comparable
CADBench
GPT-5.6 Sol86.5%SourceOrnith-1.5-35B-A3B—Not directly comparable
SpaceXAI MTS Eval
GPT-5.6 Sol52.1%SourceOrnith-1.5-35B-A3B—Not directly comparable
InferenceEval
GPT-5.6 Sol44.1%SourceOrnith-1.5-35B-A3B—Not directly comparable
KernelBench Internal
GPT-5.6 Sol73.1%SourceOrnith-1.5-35B-A3B—Not directly comparable
Terminal-Bench 2.1
GPT-5.6 Sol—Ornith-1.5-35B-A3B67.8%SourceNot directly comparable
SWE-bench Verified
GPT-5.6 Sol—Ornith-1.5-35B-A3B79%SourceNot directly comparable
SWE Multilingual
GPT-5.6 Sol—Ornith-1.5-35B-A3B71.4%SourceNot directly comparable
frontierBench
GPT-5.6 Sol—Ornith-1.5-35B-A3B5.1%SourceNot directly comparable
NL2Repo
GPT-5.6 Sol—Ornith-1.5-35B-A3B46.2%SourceNot directly comparable
Reasoning
Knowledge
GPQA
GPT-5.6 Sol94.6%SourceOrnith-1.5-35B-A3B89.2%SourceGPT-5.6 Sol leads this result
GPQA-D
GPT-5.6 Sol94.6%SourceOrnith-1.5-35B-A3B89.2%SourceGPT-5.6 Sol leads this result
HealthBench Professional
GPT-5.6 Sol60.5%SourceOrnith-1.5-35B-A3B—Not directly comparable
HealthBench Hard
GPT-5.6 Sol33.1%SourceOrnith-1.5-35B-A3B—Not directly comparable
HLE
GPT-5.6 Sol—Ornith-1.5-35B-A3B25.6%SourceNot directly comparable
HLE w/o tools
GPT-5.6 Sol—Ornith-1.5-35B-A3B25.6%SourceNot directly comparable
Math
Multimodal
Frequently asked questions
Which is better, GPT-5.6 Sol or Ornith-1.5-35B-A3B?
GPT-5.6 Sol has the higher public score, 82.04 versus 49.27, and the 90% score intervals do not overlap. That is the clearest overall quality signal in the current public data.
Which is better for coding, GPT-5.6 Sol or Ornith-1.5-35B-A3B?
The current coding averages use different weighted benchmark sets, so BenchLM does not name a winner from them. Read the shared benchmark rows directly and test the models on the same task set.
Which is better for agentic tasks, GPT-5.6 Sol or Ornith-1.5-35B-A3B?
The current agentic tasks averages use different weighted benchmark sets, so BenchLM does not name a winner from them. Read the shared benchmark rows directly and test the models on the same task set.
Which costs less, GPT-5.6 Sol or Ornith-1.5-35B-A3B?
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, GPT-5.6 Sol or Ornith-1.5-35B-A3B?
GPT-5.6 Sol has the larger documented context window: 1.05M, compared with 262K.
Related comparisons
Last updated August 19, 2026
Watch GPT-5.6 Sol vs Ornith-1.5-35B-A3B
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.