Agentic
Directional only- Qwen3.6-27B
- 30.9
- Supported · #140/152
- Qwen 3.6 Max (preview)
- 59.7
- Estimated · #22/152
- Basis
- BenchAlign lane · 6 vs 2 public rows
- Reading
- Directional only
Qwen3.6-27B vs Qwen 3.6 Max (preview)
Updated September 10, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
Qwen 3.6 Max (preview) has the higher public score, 63.7 versus 47.69, and the 90% score intervals do not overlap.
6 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
Share or export
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
Qwen3.6-27B
Qwen3.6-27B 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
Qwen 3.6 Max (preview) is 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
Qwen 3.6 Max (preview) is scored on Estimated evidence for agentic, so the reading is 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.
32 Qwen3.6-27B only
6 Shared
3 Qwen 3.6 Max (preview) only
- Shared results
- 6
- Qwen3.6-27B only
- 32
- Qwen 3.6 Max (preview) only
- 3
- Like-for-like categories
- 0 / 8
3 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Category results, on a stated basis
Each row shows the public-lane category score for both models: the BenchAlign 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.
Coding
Directional only- Qwen3.6-27B
- 41.7
- Supported · #107/151
- Qwen 3.6 Max (preview)
- 57.6
- Estimated · #32/151
- Basis
- BenchAlign lane · 6 vs 4 public rows
- Reading
- Directional only
Knowledge
Directional only- Qwen3.6-27B
- 46.6
- Estimated · #101/183
- Qwen 3.6 Max (preview)
- 60.2
- Estimated · #37/183
- Basis
- BenchAlign lane · 6 vs 1 public rows
- Reading
- Directional only
Reasoning
Not comparable- Qwen3.6-27B
- 74.1
- Unranked · 2 rankable rows
- Qwen 3.6 Max (preview)
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- Qwen3.6-27B
- 72.8
- Unranked · 5 rankable rows
- Qwen 3.6 Max (preview)
- 41.4
- Unranked · 2 rankable rows
- Basis
- Provisional lane · 2 vs 2 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Qwen3.6-27B
- Not ranked
- Qwen 3.6 Max (preview)
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- Qwen3.6-27B
- 51.9
- #35/48
- Qwen 3.6 Max (preview)
- Not ranked
- Basis
- Provisional lane · 2 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Qwen3.6-27B
- 82.4
- #51/123
- Qwen 3.6 Max (preview)
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
| Category | Qwen3.6-27B | Qwen 3.6 Max (preview) | Basis | Reading |
|---|---|---|---|---|
| Agentic | 30.9Supported · #140/152 | 59.7Estimated · #22/152 | Directional onlyBenchAlign lane · 6 vs 2 public rows | Directional only |
| Coding | 41.7Supported · #107/151 | 57.6Estimated · #32/151 | Directional onlyBenchAlign lane · 6 vs 4 public rows | Directional only |
| Knowledge | 46.6Estimated · #101/183 | 60.2Estimated · #37/183 | Directional onlyBenchAlign lane · 6 vs 1 public rows | Directional only |
| Reasoning | 74.1Unranked · 2 rankable rows | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | 72.8Unranked · 5 rankable rows | 41.4Unranked · 2 rankable rows | Not comparableProvisional lane · 2 vs 2 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multimodal | 51.9#35/48 | Not ranked | Not comparableProvisional lane · 2 vs 0 weighted rows | Not comparable |
| Instruction following | 82.4#51/123 | 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) differ from the provisional-lane categories. Unranked scores sit on the lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
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.
SuperGPQA
Knowledge
Qwen3.6-27B: 66%Qwen 3.6 Max (preview): 73.9%Normalized gap 7.9Qwen3.6-27B source Qwen 3.6 Max (preview) sourceTerminal-Bench 2.0
Agentic
Qwen3.6-27B: 59.3%Qwen 3.6 Max (preview): 65.4%Normalized gap 6.1Qwen3.6-27B source Qwen 3.6 Max (preview) sourceSWE-bench Pro
Coding
Qwen3.6-27B: 53.5%Qwen 3.6 Max (preview): 57.3%Normalized gap 3.8Qwen3.6-27B source Qwen 3.6 Max (preview) 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
- Qwen3.6-27B
- Self-hosted; infrastructure cost varies
- Fits in one request
- Qwen 3.6 Max (preview)
- API rate not published
- Fits in one request
Qwen3.6-27B has no comparable published API token rate. Qwen 3.6 Max (preview) has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- Qwen3.6-27B
- Self-hosted; infrastructure cost varies
- Fits in one request
- Qwen 3.6 Max (preview)
- API rate not published
- Fits in one request
Qwen3.6-27B has no comparable published API token rate. Qwen 3.6 Max (preview) has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Qwen3.6-27B
- Self-hosted; infrastructure cost varies
- Fits in one request
- Cached-input rate unavailable
- Qwen 3.6 Max (preview)
- API rate not published
- Fits in one request
- Cached-input rate unavailable
Qwen3.6-27B has no comparable published API token rate. Qwen 3.6 Max (preview) 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.
SpecificationQwen3.6-27BQwen 3.6 Max (preview)
Context window
Maximum documented context; output-token limits may be lower.
Qwen3.6-27B
262K
Qwen 3.6 Max (preview)
256K
API model ID
Qwen3.6-27B
Not sourced
Qwen 3.6 Max (preview)
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Qwen3.6-27B
No comparable hosted API rate
Qwen 3.6 Max (preview)
No comparable hosted API rate
Documented inputs
Qwen3.6-27B
Not sourced
Qwen 3.6 Max (preview)
Not sourced
Documented outputs
Qwen3.6-27B
Not sourced
Qwen 3.6 Max (preview)
Not sourced
Provider availability
Qwen3.6-27B
Not sourced
Qwen 3.6 Max (preview)
Not sourced
Reasoning profile
Qwen3.6-27B
Reasoning
Qwen 3.6 Max (preview)
Reasoning
Weight access
Qwen3.6-27B
Open Weight
Qwen 3.6 Max (preview)
Proprietary
License
Qwen3.6-27B
Open Weight
Qwen 3.6 Max (preview)
Proprietary
Release date
Qwen3.6-27B
2026-04-21
Qwen 3.6 Max (preview)
2026-04-20
If you already use one of these models
- Deployment change
- Both entries list Alibaba as the provider. Confirm endpoint, model ID, limits, and feature support before switching.
- Quality signal
- Qwen 3.6 Max (preview) has the higher public score, 63.7 versus 47.69, and the 90% score intervals do not overlap.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- Context tradeoff
- Qwen3.6-27B has the larger documented window (262K).
Run the same representative tasks against both endpoints before changing production traffic.
Self-host vs API cost
Estimates at 50,000 req/day · 1000 tokens/req average.
Qwen3.6-27B
API / mo$0
Self-host / mo$429
Break-even—
Qwen 3.6 Max (preview)
API / mo$0
Self-host / moNot listed
Break-even—
Proprietary model — self-hosting not applicable.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence41 rows
Agentic
Terminal-Bench 2.0
Qwen3.6-27B59.3%SourceQwen 3.6 Max (preview)65.4%SourceQwen 3.6 Max (preview) leads this result
Claw-Eval
Qwen3.6-27B72.4%SourceQwen 3.6 Max (preview)—Not directly comparable
QwenClawBench
Qwen3.6-27B53.4%SourceQwen 3.6 Max (preview)59.0%SourceQwen 3.6 Max (preview) leads this result
QwenWebBench
Qwen3.6-27B1487SourceQwen 3.6 Max (preview)—Not directly comparable
AndroidWorld
Qwen3.6-27B70.3%SourceQwen 3.6 Max (preview)—Not directly comparable
Gert Labs
Qwen3.6-27B54.84%SourceQwen 3.6 Max (preview)—Not directly comparable
Coding
SWE-bench Verified
Qwen3.6-27B77.2%SourceQwen 3.6 Max (preview)—Not directly comparable
SWE Multilingual
Qwen3.6-27B71.3%SourceQwen 3.6 Max (preview)—Not directly comparable
SWE-bench Pro
Qwen3.6-27B53.5%SourceQwen 3.6 Max (preview)57.3%SourceQwen 3.6 Max (preview) leads this result
Terminal-Bench 2.0
Qwen3.6-27B59.3%SourceQwen 3.6 Max (preview)65.4%SourceQwen 3.6 Max (preview) leads this result
LiveCodeBench
Qwen3.6-27B83.9%SourceQwen 3.6 Max (preview)—Not directly comparable
NL2Repo
Qwen3.6-27B36.2%SourceQwen 3.6 Max (preview)42.9%SourceQwen 3.6 Max (preview) leads this result
SciCode
Qwen3.6-27B—Qwen 3.6 Max (preview)47%SourceNot directly comparable
Knowledge
MMLU-Pro
Qwen3.6-27B86.2%SourceQwen 3.6 Max (preview)—Not directly comparable
MMLU-Redux
Qwen3.6-27B93.5%SourceQwen 3.6 Max (preview)—Not directly comparable
SuperGPQA
Qwen3.6-27B66%SourceQwen 3.6 Max (preview)73.9%SourceQwen 3.6 Max (preview) leads this result
C-Eval
Qwen3.6-27B91.4%SourceQwen 3.6 Max (preview)—Not directly comparable
GPQA
Qwen3.6-27B87.8%SourceQwen 3.6 Max (preview)—Not directly comparable
HLE
Qwen3.6-27B24%SourceQwen 3.6 Max (preview)—Not directly comparable
Math
HMMT Feb 2025
Qwen3.6-27B93.8%SourceQwen 3.6 Max (preview)—Not directly comparable
HMMT Nov 2025
Qwen3.6-27B90.7%SourceQwen 3.6 Max (preview)—Not directly comparable
HMMT Feb 2026
Qwen3.6-27B84.3%SourceQwen 3.6 Max (preview)—Not directly comparable
MMAnswerBench
Qwen3.6-27B80.8%SourceQwen 3.6 Max (preview)—Not directly comparable
AIME26
Qwen3.6-27B94.1%SourceQwen 3.6 Max (preview)—Not directly comparable
FrontierMath v2 (Tiers 1-3)
Qwen3.6-27B—Qwen 3.6 Max (preview)23.103%SourceNot directly comparable
FrontierMath v2 (Tier 4)
Qwen3.6-27B—Qwen 3.6 Max (preview)4.167%SourceNot directly comparable
Multimodal
MMMU
Qwen3.6-27B82.9%SourceQwen 3.6 Max (preview)—Not directly comparable
MMMU-Pro
Qwen3.6-27B75.8%SourceQwen 3.6 Max (preview)—Not directly comparable
RealWorldQA
Qwen3.6-27B84.1%SourceQwen 3.6 Max (preview)—Not directly comparable
DynaMath
Qwen3.6-27B85.6%SourceQwen 3.6 Max (preview)—Not directly comparable
MStar
Qwen3.6-27B81.4%SourceQwen 3.6 Max (preview)—Not directly comparable
SimpleVQA
Qwen3.6-27B56.1%SourceQwen 3.6 Max (preview)—Not directly comparable
CharXiv
Qwen3.6-27B78.4%SourceQwen 3.6 Max (preview)—Not directly comparable
CC-OCR
Qwen3.6-27B81.2%SourceQwen 3.6 Max (preview)—Not directly comparable
CountBench
Qwen3.6-27B97.8%SourceQwen 3.6 Max (preview)—Not directly comparable
RefCOCO (avg)
Qwen3.6-27B92.5%SourceQwen 3.6 Max (preview)—Not directly comparable
ERQA
Qwen3.6-27B62.5%SourceQwen 3.6 Max (preview)—Not directly comparable
Video-MME (with subtitle)
Qwen3.6-27B87.7%SourceQwen 3.6 Max (preview)—Not directly comparable
VideoMMMU
Qwen3.6-27B84.4%SourceQwen 3.6 Max (preview)—Not directly comparable
MLVU (M-Avg)
Qwen3.6-27B86.6%SourceQwen 3.6 Max (preview)—Not directly comparable
V*
Qwen3.6-27B94.7%SourceQwen 3.6 Max (preview)—Not directly comparable
Frequently asked questions
Which is better, Qwen3.6-27B or Qwen 3.6 Max (preview)?
Qwen 3.6 Max (preview) has the higher public score, 63.7 versus 47.69, 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, Qwen3.6-27B or Qwen 3.6 Max (preview)?
Qwen 3.6 Max (preview) scores higher for coding on the public lane, 57.6 to 41.7. Qwen 3.6 Max (preview) is 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, Qwen3.6-27B or Qwen 3.6 Max (preview)?
Qwen 3.6 Max (preview) scores higher for agentic tasks on the public lane, 59.7 to 30.9. Qwen 3.6 Max (preview) is 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, Qwen3.6-27B or Qwen 3.6 Max (preview)?
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, Qwen3.6-27B or Qwen 3.6 Max (preview)?
Qwen3.6-27B has the larger documented context window: 262K, compared with 256K.
Related comparisons
Last updated September 10, 2026
Watch Qwen3.6-27B vs Qwen 3.6 Max (preview)
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.