Agentic
Like-for-like- Inkling-Small
- 36.9
- Supported · #128/152
- Qwen3.7 Max
- 41.2
- Supported · #110/152
- Basis
- BenchAlign lane · 5 vs 10 public rows
- Reading
- Qwen3.7 Max leads · intervals overlap
Inkling-Small vs Qwen3.7 Max
Updated September 10, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
Qwen3.7 Max has the higher public score estimate, 67.08 versus 59.25, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.
17 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.
Coding work
Code generation, repair, and software-engineering tasks
Qwen3.7 Max
Qwen3.7 Max leads on the public coding lane, 49.3 to 43.8, with Supported evidence for both models, although the 90% intervals overlap.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Qwen3.7 Max
Qwen3.7 Max leads on the public agentic lane, 41.2 to 36.9, with Supported evidence for both models, although the 90% intervals overlap.
Confidence: limited
Show secondary and unsupported calls
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
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
8 Inkling-Small only
17 Shared
24 Qwen3.7 Max only
- Shared results
- 17
- Inkling-Small only
- 8
- Qwen3.7 Max only
- 24
- Like-for-like categories
- 4 / 8
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
Like-for-like- Inkling-Small
- 43.8
- Supported · #97/151
- Qwen3.7 Max
- 49.3
- Supported · #64/151
- Basis
- BenchAlign lane · 6 vs 10 public rows
- Reading
- Qwen3.7 Max leads · intervals overlap
Knowledge
Like-for-like- Inkling-Small
- 56.5
- Supported · #46/183
- Qwen3.7 Max
- 61.7
- Supported · #28/183
- Basis
- BenchAlign lane · 6 vs 9 public rows
- Reading
- Qwen3.7 Max leads · intervals overlap
Instruction following
Like-for-like- Inkling-Small
- 89.6
- #25/123
- Qwen3.7 Max
- 91.1
- #16/123
- Basis
- Provisional lane · 1 vs 1 weighted rows
- Reading
- Qwen3.7 Max leads
Reasoning
Not comparable- Inkling-Small
- 42.7
- Unranked · 3 rankable rows
- Qwen3.7 Max
- 75.1
- Unranked · 3 rankable rows
- Basis
- Provisional lane · 1 vs 1 weighted rows
- Reading
- Not comparable
Math
Not comparable- Inkling-Small
- 76.9
- Unranked · 2 rankable rows
- Qwen3.7 Max
- 82.1
- Unranked · 3 rankable rows
- Basis
- Provisional lane · 2 vs 1 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Inkling-Small
- Not ranked
- Qwen3.7 Max
- 100.0
- #1/12
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- Inkling-Small
- 48.8
- #39/48
- Qwen3.7 Max
- Not ranked
- Basis
- Provisional lane · 2 vs 0 weighted rows
- Reading
- Not comparable
| Category | Inkling-Small | Qwen3.7 Max | Basis | Reading |
|---|---|---|---|---|
| Agentic | 36.9Supported · #128/152 | 41.2Supported · #110/152 | Like-for-likeBenchAlign lane · 5 vs 10 public rows | Qwen3.7 Max leads · intervals overlap |
| Coding | 43.8Supported · #97/151 | 49.3Supported · #64/151 | Like-for-likeBenchAlign lane · 6 vs 10 public rows | Qwen3.7 Max leads · intervals overlap |
| Knowledge | 56.5Supported · #46/183 | 61.7Supported · #28/183 | Like-for-likeBenchAlign lane · 6 vs 9 public rows | Qwen3.7 Max leads · intervals overlap |
| Instruction following | 89.6#25/123 | 91.1#16/123 | Like-for-likeProvisional lane · 1 vs 1 weighted rows | Qwen3.7 Max leads |
| Reasoning | 42.7Unranked · 3 rankable rows | 75.1Unranked · 3 rankable rows | Not comparableProvisional lane · 1 vs 1 weighted rows | Not comparable |
| Math | 76.9Unranked · 2 rankable rows | 82.1Unranked · 3 rankable rows | Not comparableProvisional lane · 2 vs 1 weighted rows | Not comparable |
| Multilingual | Not ranked | 100.0#1/12 | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
| Multimodal | 48.8#39/48 | Not ranked | Not comparableProvisional lane · 2 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.
HMMT Feb 2026
Math
HLE
Knowledge
Terminal-Bench 2.0
Agentic
SciCode
Coding
SWE-bench Pro
Coding
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
- Inkling-Small
- $0.0013
- Fits in one request
- Qwen3.7 Max
- API rate not published
- Fits in one request
Qwen3.7 Max has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- Inkling-Small
- $0.03332
- Fits in one request
- Qwen3.7 Max
- API rate not published
- Fits in one request
Qwen3.7 Max has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Inkling-Small
- $0.0492
- Fits in one request
- Qwen3.7 Max
- API rate not published
- Fits in one request
- Cached-input rate unavailable
Qwen3.7 Max 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.
SpecificationInkling-SmallQwen3.7 Max
Context window
Maximum documented context; output-token limits may be lower.
Inkling-Small
1M
Qwen3.7 Max
1M
API model ID
Inkling-Small
Not sourced
Qwen3.7 Max
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Inkling-Small
$0.116 per 1M cached input tokens
Qwen3.7 Max
No comparable hosted API rate
Documented inputs
Inkling-Small
Not sourced
Qwen3.7 Max
Not sourced
Documented outputs
Inkling-Small
Not sourced
Qwen3.7 Max
Not sourced
Provider availability
Inkling-Small
Not sourced
Qwen3.7 Max
Not sourced
Reasoning profile
Inkling-Small
Hybrid
Qwen3.7 Max
Reasoning
Weight access
Inkling-Small
Open Weight
Qwen3.7 Max
Proprietary
License
Inkling-Small
Open Weight
Qwen3.7 Max
Proprietary
Release date
Inkling-Small
2026-07-30
Qwen3.7 Max
2026-05-16
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
- Qwen3.7 Max has the higher public score estimate, 67.08 versus 59.25, but the 90% score intervals overlap.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- Context tradeoff
- Both models list 1M.
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 evidence49 rows
Agentic
Terminal-Bench 2.0
Inkling-Small64.7%SourceQwen3.7 Max69.7%SourceQwen3.7 Max leads this result
BrowseComp
Inkling-Small77.4%SourceQwen3.7 Max—Not directly comparable
MCP Atlas
Inkling-Small79.6%SourceQwen3.7 Max76.4%SourceInkling-Small leads this result
Toolathlon-Verified
Inkling-Small54.4%SourceQwen3.7 Max—Not directly comparable
Terminal-Bench 2.1 (Vals)
Inkling-Small55.1%SourceQwen3.7 Max61.0%SourceQwen3.7 Max leads this result
QwenClawBench
Inkling-Small—Qwen3.7 Max64.3%SourceNot directly comparable
Claw-Eval
Inkling-Small—Qwen3.7 Max65.2%SourceNot directly comparable
BFCL v4
Inkling-Small—Qwen3.7 Max75.0%SourceNot directly comparable
VITA-Bench
Inkling-Small—Qwen3.7 Max47.9%SourceNot directly comparable
HLE w/ tools
Inkling-Small—Qwen3.7 Max53.5%SourceNot directly comparable
Gert Labs
Inkling-Small—Qwen3.7 Max64.27%SourceNot directly comparable
ResearchClawBench
Inkling-Small—Qwen3.7 Max18.7%SourceNot directly comparable
Coding
SWE-bench Verified
Inkling-Small80.2%SourceQwen3.7 Max80.4%SourceQwen3.7 Max leads this result
SWE-bench Pro
Inkling-Small55.9%SourceQwen3.7 Max60.6%SourceQwen3.7 Max leads this result
Terminal-Bench 2.0
Inkling-Small64.7%SourceQwen3.7 Max69.7%SourceQwen3.7 Max leads this result
SciCode
Inkling-Small48.7%SourceQwen3.7 Max53.5%SourceQwen3.7 Max leads this result
LiveCodeBench (Vals)
Inkling-Small85.9%SourceQwen3.7 Max87.1%SourceQwen3.7 Max leads this result
SWE-bench (Vals)
Inkling-Small82.2%SourceQwen3.7 Max68.8%SourceInkling-Small leads this result
SWE Multilingual
Inkling-Small—Qwen3.7 Max78.3%SourceNot directly comparable
NL2Repo
Inkling-Small—Qwen3.7 Max47.2%SourceNot directly comparable
LiveCodeBench
Inkling-Small—Qwen3.7 Max91.6%SourceNot directly comparable
OpenHarmony Bench
Inkling-Small—Qwen3.7 Max53.4%SourceNot directly comparable
Reasoning
Knowledge
GPQA
Inkling-Small89.5%SourceQwen3.7 Max92.4%SourceQwen3.7 Max leads this result
GPQA-D
Inkling-Small89.5%SourceQwen3.7 Max92.4%SourceQwen3.7 Max leads this result
HLE
Inkling-Small47.8%SourceQwen3.7 Max41.4%SourceInkling-Small leads this result
HLE w/o tools
Inkling-Small31.6%SourceQwen3.7 Max—Not directly comparable
GPQA Diamond (Vals)
Inkling-Small83.6%SourceQwen3.7 Max90.2%SourceQwen3.7 Max leads this result
MMLU-Pro (Vals)
Inkling-Small85.6%SourceQwen3.7 Max89.3%SourceQwen3.7 Max leads this result
MMLU-Pro
Inkling-Small—Qwen3.7 Max89.6%SourceNot directly comparable
MMLU-Redux
Inkling-Small—Qwen3.7 Max95%SourceNot directly comparable
SuperGPQA
Inkling-Small—Qwen3.7 Max73.6%SourceNot directly comparable
MMMLU
Inkling-Small—Qwen3.7 Max90.3%SourceNot directly comparable
Math
Multilingual
MMLU-ProX
Inkling-Small—Qwen3.7 Max87%SourceNot directly comparable
NOVA-63
Inkling-Small—Qwen3.7 Max59.0%SourceNot directly comparable
INCLUDE
Inkling-Small—Qwen3.7 Max86.2%SourceNot directly comparable
MAXIFE
Inkling-Small—Qwen3.7 Max89.2%SourceNot directly comparable
PolyMath
Inkling-Small—Qwen3.7 Max86.5%SourceNot directly comparable
Multimodal
Instruction following
Frequently asked questions
Which is better, Inkling-Small or Qwen3.7 Max?
Qwen3.7 Max has the higher public score estimate, 67.08 versus 59.25, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Which is better for coding, Inkling-Small or Qwen3.7 Max?
Qwen3.7 Max leads the public coding lane, 49.3 to 43.8, with Supported evidence for both models, although the 90% intervals overlap.
Which is better for agentic tasks, Inkling-Small or Qwen3.7 Max?
Qwen3.7 Max leads the public agentic tasks lane, 41.2 to 36.9, with Supported evidence for both models, although the 90% intervals overlap.
Which costs less, Inkling-Small or Qwen3.7 Max?
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, Inkling-Small or Qwen3.7 Max?
Both models list the same context window, 1M.
Related comparisons
Last updated September 10, 2026
Watch Inkling-Small vs Qwen3.7 Max
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.