Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
Bev / Bonsai 27B is not ranked on the public lane for coding, so no winner is named for coding.
Confidence: limited
Bev / Bonsai 27B vs Claude Opus 5
Updated October 1, 2026. We do not rank this pair: at least one has no public score. Public scores include evidence status and uncertainty.
Both of these models will change. Get the price, version and retirement notices for the pair, each with its source. Follow model changes
Share or export
Decision reading
The public evidence has no benchmark result shared by both models, so it does not support a quality verdict. Use the documented cost, context, and runtime rows instead. 0 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
- Shared results
- 0
- Bev / Bonsai 27B only
- 1
- Claude Opus 5 only
- 77
- Like-for-like categories
- 0 / 8
Supported: Claude Opus 5How 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
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
Bev / Bonsai 27B is not ranked on the public lane for agentic, so no winner is named for agentic.
Confidence: limitedLong documents
Prompts that approach the documented context limit
Not enough matched evidence
A complete context comparison is not sourced.
Confidence: limitedChat 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-ratesCache-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-ratesRepository 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.
—Bev / Bonsai 27B72.1Claude Opus 5
Not comparable · BenchAlign v5.8
The coding row is not comparable on the public lane: at least one model is not measured or not ranked there.
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.
1 Bev / Bonsai 27B only
77 Claude Opus 5 only
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.
A shared-evidence shape is not available.
BenchLM does not draw a radar or infer missing axes when the matched evidence is too sparse.
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
Not comparable- Bev / Bonsai 27B
- Not ranked
- Claude Opus 5
- 77.6
- Supported · #3/119
- Basis
- BenchAlign v5.8 lane · 0 vs 20 public rows
- Reading
- Not comparable
Coding
Not comparable- Bev / Bonsai 27B
- Not ranked
- Claude Opus 5
- 72.1
- Supported · #7/144
- Basis
- BenchAlign v5.8 lane · 0 vs 18 public rows
- Reading
- Not comparable
Reasoning
Not comparable- Bev / Bonsai 27B
- Not ranked
- Claude Opus 5
- 77.3
- #10/27
- Basis
- Provisional lane · 0 vs 2 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- Bev / Bonsai 27B
- Not ranked
- Claude Opus 5
- 88.8
- #2/49
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
Knowledge
Not comparable- Bev / Bonsai 27B
- Not ranked
- Claude Opus 5
- 80.2
- Supported · #7/170
- Basis
- BenchAlign v5.8 lane · 0 vs 19 public rows
- Reading
- Not comparable
Multilingual
Not comparable- Bev / Bonsai 27B
- Not ranked
- Claude Opus 5
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Bev / Bonsai 27B
- Not ranked
- Claude Opus 5
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- Bev / Bonsai 27B
- Not ranked
- Claude Opus 5
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
| Category | Bev / Bonsai 27B | Claude Opus 5 | Basis | Reading |
|---|---|---|---|---|
| Agentic | Not ranked | 77.6Supported · #3/119 | Not comparableBenchAlign v5.8 lane · 0 vs 20 public rows | Not comparable |
| Coding | Not ranked | 72.1Supported · #7/144 | Not comparableBenchAlign v5.8 lane · 0 vs 18 public rows | Not comparable |
| Reasoning | Not ranked | 77.3#10/27 | Not comparableProvisional lane · 0 vs 2 weighted rows | Not comparable |
| Multimodal | Not ranked | 88.8#2/49 | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
| Knowledge | Not ranked | 80.2Supported · #7/170 | Not comparableBenchAlign v5.8 lane · 0 vs 19 public rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Instruction following | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | 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
- Bev / Bonsai 27B
- API rate not published
- Fit state unavailable
- Claude Opus 5
- $0.0175
- Fits in one request
Bev / Bonsai 27B has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- Bev / Bonsai 27B
- API rate not published
- Fit state unavailable
- Claude Opus 5
- $0.325
- Fits in one request
Bev / Bonsai 27B has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Bev / Bonsai 27B
- API rate not published
- Fit state unavailable
- Cached-input rate unavailable
- Claude Opus 5
- $0.45
- Fits in one request
Bev / Bonsai 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.
SpecificationBev / Bonsai 27BClaude Opus 5
Context window
Maximum documented context; output-token limits may be lower.
Bev / Bonsai 27B
Not sourced
Claude Opus 5
API model ID
Bev / Bonsai 27B
Not sourced
Claude Opus 5
claude-opus-5
Anthropic model overviewCached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Bev / Bonsai 27B
No comparable hosted API rate
Provider pricingClaude Opus 5
$0.5 per 1M cached input tokens
Claude API pricingDocumented inputs
Bev / Bonsai 27B
Not sourced
Claude Opus 5
text, image
Anthropic model overviewDocumented outputs
Bev / Bonsai 27B
Not sourced
Claude Opus 5
Provider availability
Bev / Bonsai 27B
Not sourced
Claude Opus 5
Generally Available · Claude API
Anthropic model overviewReasoning profile
Bev / Bonsai 27B
Non-Reasoning
Claude Opus 5
Reasoning
Weight access
Bev / Bonsai 27B
Open Weight
Claude Opus 5
Proprietary
License
Bev / Bonsai 27B
Open Weight
Claude Opus 5
Proprietary
Release date
Bev / Bonsai 27B
Not sourced
Claude Opus 5
2026-07-24
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
- The public evidence has no benchmark result shared by both models, so it does not support a quality verdict.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- Context tradeoff
- A complete documented context comparison is not available.
Run the same representative tasks against both endpoints before changing production traffic.
Questions
Which is better, Bev / Bonsai 27B or Claude Opus 5?
The public evidence has no benchmark result shared by both models, so it does not support a quality verdict. The page therefore keeps the decision tied to the specific documented workload.
Which is better for coding, Bev / Bonsai 27B or Claude Opus 5?
Bev / Bonsai 27B is not ranked on the public lane for coding, so no winner is named for coding.
Which is better for agentic tasks, Bev / Bonsai 27B or Claude Opus 5?
Bev / Bonsai 27B is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.
Which costs less, Bev / Bonsai 27B or Claude Opus 5?
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, Bev / Bonsai 27B or Claude Opus 5?
A complete documented context-window comparison is not available.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence78 rows
Agentic
Terminal-Bench 3.0
Bev / Bonsai 27B—Claude Opus 542.7%SourceNot directly comparable
BrowseComp
Bev / Bonsai 27B—Claude Opus 590.8%SourceNot directly comparable
HLE w/ tools
Bev / Bonsai 27B—Claude Opus 564.7%SourceNot directly comparable
DeepSearchQA
Bev / Bonsai 27B—Claude Opus 595.0%SourceNot directly comparable
DRACO
Bev / Bonsai 27B—Claude Opus 588.6%SourceNot directly comparable
BrowseComp (10-agent, prerelease)
Bev / Bonsai 27B—Claude Opus 593.6%SourceNot directly comparable
OSWorld 2.0
Bev / Bonsai 27B—Claude Opus 570.6%SourceNot directly comparable
MCP Atlas
Bev / Bonsai 27B—Claude Opus 585.8%SourceNot directly comparable
MCP-Atlas claim coverage
Bev / Bonsai 27B—Claude Opus 589.1%SourceNot directly comparable
LAB all-pass (Anthropic harness)
Bev / Bonsai 27B—Claude Opus 523.58%SourceNot directly comparable
LAB criterion-pass (Anthropic harness)
Bev / Bonsai 27B—Claude Opus 593.74%SourceNot directly comparable
LAB all-pass (Harvey held-out)
Bev / Bonsai 27B—Claude Opus 511.7%SourceNot directly comparable
LAB criterion-pass (Harvey held-out)
Bev / Bonsai 27B—Claude Opus 594.1%SourceNot directly comparable
Toolathlon-Verified
Bev / Bonsai 27B—Claude Opus 580.6%SourceNot directly comparable
Toolathlon Verified Pass@3
Bev / Bonsai 27B—Claude Opus 587.0%SourceNot directly comparable
Toolathlon Verified Pass³
Bev / Bonsai 27B—Claude Opus 573.1%SourceNot directly comparable
Toolathlon Verified avg. turns
Bev / Bonsai 27B—Claude Opus 523.5 turnsSourceNot directly comparable
AutomationBench
Bev / Bonsai 27B—Claude Opus 526.0%SourceNot directly comparable
Terminal-Bench 2.1 (Vals)
Bev / Bonsai 27B—Claude Opus 584.6%SourceNot directly comparable
ApprenticeBench
Bev / Bonsai 27B—Claude Opus 536%SourceNot directly comparable
Coding
Bug Hunt Bench
Bev / Bonsai 27B—Claude Opus 527 fixesSourceNot directly comparable
SWE-bench Verified
Bev / Bonsai 27B—Claude Opus 596%SourceNot directly comparable
SWE-bench Pro
Bev / Bonsai 27B—Claude Opus 579.2%SourceNot directly comparable
SWE Multilingual
Bev / Bonsai 27B—Claude Opus 589.5%SourceNot directly comparable
SWE Multimodal
Bev / Bonsai 27B—Claude Opus 559.4%SourceNot directly comparable
DeepSWE
Bev / Bonsai 27B—Claude Opus 568.8%SourceNot directly comparable
FrontierCode 1.1 Main
Bev / Bonsai 27B—Claude Opus 553.4%SourceNot directly comparable
FrontierCode 1.1 Extended
Bev / Bonsai 27B—Claude Opus 563.6%SourceNot directly comparable
FrontierSWE v2
Bev / Bonsai 27B—Claude Opus 552.0%SourceNot directly comparable
ProgramBench (episode 1)
Bev / Bonsai 27B—Claude Opus 583.0%SourceNot directly comparable
ProgramBench
Bev / Bonsai 27B—Claude Opus 593.0%SourceNot directly comparable
CursorBench 3.2
Bev / Bonsai 27B—Claude Opus 570.0%SourceNot directly comparable
VulcanBench v3
Bev / Bonsai 27B—Claude Opus 587.0%SourceNot directly comparable
VulcanBench CII v1
Bev / Bonsai 27B—Claude Opus 596.4%SourceNot directly comparable
LiveCodeBench (Vals)
Bev / Bonsai 27B—Claude Opus 589.0%SourceNot directly comparable
SWE-bench (Vals)
Bev / Bonsai 27B—Claude Opus 597.0%SourceNot directly comparable
CursorBench 4.0
Bev / Bonsai 27B—Claude Opus 546.6%SourceNot directly comparable
PostTrainBench v1.1
Bev / Bonsai 27B—Claude Opus 535.0%SourceNot directly comparable
Reasoning
JevBench 1.5
Bev / Bonsai 27B15.77SourceClaude Opus 5—Not directly comparable
ARC-AGI-1
Bev / Bonsai 27B—Claude Opus 597.50%SourceNot directly comparable
ARC-AGI-2
Bev / Bonsai 27B—Claude Opus 590.4%SourceNot directly comparable
ARC-AGI-3
Bev / Bonsai 27B—Claude Opus 530.2%SourceNot directly comparable
Multimodal
Chartography (no tools)
Bev / Bonsai 27B—Claude Opus 529.6%SourceNot directly comparable
Chartography (tools)
Bev / Bonsai 27B—Claude Opus 583.0%SourceNot directly comparable
BenchCAD Vision2Code (no tools)
Bev / Bonsai 27B—Claude Opus 50.366SourceNot directly comparable
BenchCAD Vision2Code (tools)
Bev / Bonsai 27B—Claude Opus 50.821SourceNot directly comparable
GDP.pdf (no tools)
Bev / Bonsai 27B—Claude Opus 583.4%SourceNot directly comparable
GDP.pdf (tools)
Bev / Bonsai 27B—Claude Opus 585.5%SourceNot directly comparable
OfficeQA
Bev / Bonsai 27B—Claude Opus 578.1%SourceNot directly comparable
OfficeQA Pro
Bev / Bonsai 27B—Claude Opus 566.9%SourceNot directly comparable
Knowledge
HLE
Bev / Bonsai 27B—Claude Opus 564.7%SourceNot directly comparable
HLE w/o tools
Bev / Bonsai 27B—Claude Opus 556.3%SourceNot directly comparable
HLE-Verified
Bev / Bonsai 27B—Claude Opus 554.4%SourceNot directly comparable
LABBench2
Bev / Bonsai 27B—Claude Opus 584.2%SourceNot directly comparable
HealthBench (raw)
Bev / Bonsai 27B—Claude Opus 567.1%SourceNot directly comparable
HealthBench (length-adjusted)
Bev / Bonsai 27B—Claude Opus 557.8%SourceNot directly comparable
HealthBench Professional
Bev / Bonsai 27B—Claude Opus 559.8%SourceNot directly comparable
HealthBench Professional (raw)
Bev / Bonsai 27B—Claude Opus 573.4%SourceNot directly comparable
BioMysteryBench (human-solvable)
Bev / Bonsai 27B—Claude Opus 590.1%SourceNot directly comparable
BioMysteryBench (human-difficult)
Bev / Bonsai 27B—Claude Opus 549.4%SourceNot directly comparable
SpatialBench Verified
Bev / Bonsai 27B—Claude Opus 572.5%SourceNot directly comparable
SingleCellBench
Bev / Bonsai 27B—Claude Opus 560.6%SourceNot directly comparable
ProteinGym Hard
Bev / Bonsai 27B—Claude Opus 547.7%SourceNot directly comparable
Protein Design
Bev / Bonsai 27B—Claude Opus 542.5%SourceNot directly comparable
Organic chemistry V2
Bev / Bonsai 27B—Claude Opus 561.6%SourceNot directly comparable
Protocols (troubleshooting)
Bev / Bonsai 27B—Claude Opus 561.1%SourceNot directly comparable
Protocols (understanding)
Bev / Bonsai 27B—Claude Opus 578.4%SourceNot directly comparable
GPQA Diamond (Vals)
Bev / Bonsai 27B—Claude Opus 593.4%SourceNot directly comparable
MMLU-Pro (Vals)
Bev / Bonsai 27B—Claude Opus 591.6%SourceNot directly comparable
Multilingual
Instruction following
Gray Swan IPI (15 attempts)
Bev / Bonsai 27B—Claude Opus 54.6%SourceNot directly comparable
Math
IMO 2026
Bev / Bonsai 27B—Claude Opus 542/42SourceNot directly comparable
RiemannBench (no tools)
Bev / Bonsai 27B—Claude Opus 560.0%SourceNot directly comparable
RiemannBench (tools)
Bev / Bonsai 27B—Claude Opus 579.0%SourceNot directly comparable
ArXivMath Jun. 2026 (no tools)
Bev / Bonsai 27B—Claude Opus 590.8%SourceNot directly comparable
ArXivMath Jun. 2026 (tools)
Bev / Bonsai 27B—Claude Opus 591.3%SourceNot directly comparable
78 public results · 0 shared
Watch Bev / Bonsai 27B vs Claude Opus 5
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.
Last updated October 1, 2026