Coding work
Code generation, repair, and software-engineering tasks
Claude Sonnet 4.6
Claude Sonnet 4.6 leads on the public coding lane, 48.1 to 35.7, with Supported evidence for both models, although the 90% intervals overlap.
Updated September 27, 2026. Rank says Claude Sonnet 4.6 is ahead. Price, access, and your workload can each overturn that. 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
Claude Sonnet 4.6 has the higher public score estimate, 56.35 versus 40.06, but the 90% score intervals overlap. Treat that as a lead, not a settled winner. 7 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
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.
Code generation, repair, and software-engineering tasks
Claude Sonnet 4.6
Claude Sonnet 4.6 leads on the public coding lane, 48.1 to 35.7, with Supported evidence for both models, although the 90% intervals overlap.
Prompts that approach the documented context limit
Qwen3.5-122B-A10B
Qwen3.5-122B-A10B has the larger documented context window.
Tool use, computer use, and multi-step task completion
Not enough matched evidence
Qwen3.5-122B-A10B is scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.
1K fresh input + 500 output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
The page does not recommend a cost winner because at least one model cannot fit the stated workload in one request. Claude Sonnet 4.6 does not fit this workload in one request. Claude Sonnet 4.6 has no published cached-input rate, so cached tokens use its listed input rate. Qwen3.5-122B-A10B has no comparable published API token rate.
50K fresh input + 3K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
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.
Like-for-like · BenchAlign v5.7
Claude Sonnet 4.6 leads the like-for-like coding row, although the 90% intervals overlap.
Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
2 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
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.
SuperGPQAKnowledge
Normalized gap 27.9OSWorld-VerifiedAgentic
Normalized gap 14.1Terminal-Bench 2.0Agentic
Normalized gap 9.7SWE-bench VerifiedCoding
Normalized gap 7.6MMLU-ProKnowledge
Normalized gap 7.5Each row shows the public-lane category score for both models: the BenchAlign v5.7 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.
| Category | Claude Sonnet 4.6 | Qwen3.5-122B-A10B | Basis | Reading |
|---|---|---|---|---|
| Coding | 48.1Supported · #41/135 | 35.7Supported · #73/135 | Like-for-likeBenchAlign v5.7 lane · 8 vs 1 public rows | Claude Sonnet 4.6 leads · intervals overlap |
| Multimodal | 54.1#35/50 | 57.0#34/50 | Like-for-likeProvisional lane · 1 vs 1 weighted rows | Qwen3.5-122B-A10B leads |
| Knowledge | 53.0Supported · #48/158 | 41.5Supported · #86/158 | Like-for-likeBenchAlign v5.7 lane · 6 vs 3 public rows | Claude Sonnet 4.6 leads · intervals overlap |
| Agentic | 41.8Supported · #44/105 | 23.1Estimated · #84/105 | Directional onlyBenchAlign v5.7 lane · 9 vs 3 public rows | Directional only |
| Instruction following | 46.6#88/124 | 91.6#11/124 | Directional onlyProvisional lane · 0 vs 0 weighted rows | Directional only |
| Reasoning | 69.0Unranked · 2 rankable rows | 49.8Unranked · 3 rankable rows | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
| Multilingual | Not ranked | 36.8#10/12 | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
| Math | 48.8Unranked · 2 rankable rows | 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 v5.7) 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.
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.
1K fresh input + 500 output tokens
Qwen3.5-122B-A10B has no comparable published API token rate.
50K fresh input + 3K output tokens
Qwen3.5-122B-A10B has no comparable published API token rate.
200K cached + 20K fresh input + 10K output tokens
Claude Sonnet 4.6 does not fit this workload in one request. Claude Sonnet 4.6 has no published cached-input rate, so cached tokens use its listed input rate. Qwen3.5-122B-A10B has no comparable published API token rate.
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
Maximum documented context; output-token limits may be lower.
Claude Sonnet 4.6
200K
Qwen3.5-122B-A10B
262K
Claude Sonnet 4.6
Not sourced
Qwen3.5-122B-A10B
Not sourced
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Claude Sonnet 4.6
Not published
Qwen3.5-122B-A10B
No comparable hosted API rate
Claude Sonnet 4.6
Not sourced
Qwen3.5-122B-A10B
Not sourced
Claude Sonnet 4.6
Not sourced
Qwen3.5-122B-A10B
Not sourced
Claude Sonnet 4.6
Not sourced
Qwen3.5-122B-A10B
Not sourced
Claude Sonnet 4.6
Non-Reasoning
Qwen3.5-122B-A10B
Reasoning
Claude Sonnet 4.6
Proprietary
Qwen3.5-122B-A10B
Open Weight
Claude Sonnet 4.6
Proprietary
Qwen3.5-122B-A10B
Open Weight
Claude Sonnet 4.6
2026-02-01
Qwen3.5-122B-A10B
2026-03-04
Claude Sonnet 4.6 has the higher public score estimate, 56.35 versus 40.06, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Claude Sonnet 4.6 leads the public coding lane, 48.1 to 35.7, with Supported evidence for both models, although the 90% intervals overlap.
Claude Sonnet 4.6 scores higher for agentic tasks on the public lane, 41.8 to 23.1. Qwen3.5-122B-A10B 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.
Both models do not have comparable published API token rates, so this page does not name a universal price winner.
Qwen3.5-122B-A10B has the larger documented context window: 262K, compared with 200K.
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Terminal-Bench 2.0
Claude Sonnet 4.6 leads this result
OSWorld-Verified
Claude Sonnet 4.6 leads this result
Claw-Eval
Not directly comparable
CyberGym
Not directly comparable
Gert Labs
Not directly comparable
OSWorld 2.0
Not directly comparable
JobBench
Not directly comparable
Terminal-Bench 2.1 (Vals)
Not directly comparable
ApprenticeBench
Not directly comparable
BrowseComp
Not directly comparable
SWE-bench Verified
Claude Sonnet 4.6 leads this result
SWE-Rebench
Not directly comparable
React Native Evals
Not directly comparable
Vibe Code Bench
Not directly comparable
cursorBench31
Not directly comparable
FrontierCode 1.1 Main
Not directly comparable
LiveCodeBench (Vals)
Not directly comparable
SWE-bench (Vals)
Not directly comparable
LongBench v2
Not directly comparable
CharXiv
Claude Sonnet 4.6 leads this result
MMMU
Not directly comparable
MMVU
Not directly comparable
MathVision
Not directly comparable
V*
Not directly comparable
GPQA
Claude Sonnet 4.6 leads this result
SuperGPQA
Claude Sonnet 4.6 leads this result
MMLU-Pro
Qwen3.5-122B-A10B leads this result
HLE
Not directly comparable
GPQA Diamond (Vals)
Not directly comparable
MMLU-Pro (Vals)
Not directly comparable
MMLU-ProX
Not directly comparable
IFEval
Not directly comparable
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 September 27, 2026