Skip to main content
Radar

Five or fewer confirmed AI changes, with original sources, on mornings when something changed.A free source-linked morning brief.

Start free brief

Qwen3.5-122B-A10B

CurrentReleased Mar 4, 2026Open WeightReasoning262K context

Released Mar 4, 2026 see all recent releases

Decision reading
Qwen3.5-122B-A10B scores 59.5 out of 100 and ranks #65 of 217. This profile shows 15 source-displayable benchmark rows; its strongest eligible category is Multilingual at #9. Published weights can be self-hosted, but infrastructure cost varies and is not a comparable API token rate.

Data as of August 13, 2026 · How the score is built

Strongest published evidence

Multilingual ranks #9. A well-rounded choice across a range of tasks.

Validate before choosing

15 published rows leave some tracked benchmark slots empty. No comparable first-party API token rate is published.

Decision snapshot

Each value carries a field reference instead of floating alone. Markers compare this model with the current ranked and priced catalog; they are not absolute quality thresholds.

Capability

59.5/100

field median 57.7

#65 of 217 ranked models

Price

Self-hosted; infrastructure cost varies

input median $1

No comparable first-party hosted token rate

Speed

133tok/s

field median 93 tok/s

First token 17.32 s

Context

262Ktokens

field median 200,000

Reported for this model; direct source link not stored

Capability shape

Each axis shows percentile within that category’s eligible cohort. The comparison outline is the median of the six nearest public-score peers; a collapsed vertex means the category is not rank-eligible.

Qwen3.5-122B-A10B category percentile values

  • Agentic52nd percentile
  • Coding41st percentile
  • ReasoningNot eligible
  • Knowledge74th percentile
  • MathNot eligible
  • Multilingual33rd percentile
  • MultimodalNot eligible
  • Instruction following68th percentile

The dashed outline is median of 6 nearest peers.

Eligible category ranks

  1. Agentic#63/129
  2. Coding#80/134
  3. ReasoningNot ranked
  4. Knowledge#15/55
  5. MathNot ranked
  6. Multilingual#9/13
  7. MultimodalNot ranked
  8. Inst. Following#13/38
Top decileTop quartileMid-fieldNot eligible

How much of this is verified

Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.

  1. Agentic3/3 verified
  2. Coding1/1 verified
  3. Reasoning1/1 verified
  4. Knowledge3/3 verified
  5. MathNot measured
  6. Multilingual1/1 verified
  7. Multimodal3/5 verified
  8. Inst. Following1/1 verified
Verified sourceProvisionalNot measured

Spec sheet

Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.

API model ID
Not published
Context window
262K
Maximum output
Not sourced yet
Knowledge cutoff
Not sourced yet
Input modalities
Not sourced yet
Output modalities
Not sourced yet
Parameters
Not sourced yet
Availability
Not sourced yet
Cloud regions
Not tracked yet
Lifecycle
Current
API capabilities
Tool calling, structured outputs, and batch support are not tracked yet
Prompt caching
Not documented in the pricing record
Self-host
Open weights available; hardware estimate not sourced
Rate limits
Not tracked yet

Deployment options

Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.

Published weights are available, but BenchLM does not yet have a sourced parameter and VRAM profile for this exact model. Hardware cost estimates stay unavailable until that sizing record is complete.

Estimate VRAM from known parameters

Category score record

Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.

Category scores, ranks, weighting, benchmark coverage, and evidence status
CategoryScore
AgenticRank #63 of 129Percentile 52ndWeight 22%3 benchmarksVerified48.6
CodingRank #80 of 134Percentile 41stWeight 20%1 benchmarkVerified47.4
ReasoningRank Not rankedWeight 17%1 benchmarkVerified74.3
KnowledgeRank #15 of 55Percentile 74thWeight 12%3 benchmarksVerified77.7
MathWeight 5%0 benchmarksNot measuredNot measured
MultilingualRank #9 of 13Percentile 33rdWeight 7%1 benchmarkVerified36.8
MultimodalRank Not rankedWeight 12%5 benchmarksMixed sources54.8
Inst. FollowingRank #13 of 38Percentile 68thWeight 5%1 benchmarkVerified88.4

Benchmark ledger

Coding opens by default. The marker compares each value with the best source-verified result in the catalog; provisional leaders do not set the reference. Expand the remaining categories for every published row.

Coding1 row
Coding benchmark values, best verified comparison, weight, and source status
SWE-bench VerifiedSoftware Engineering Benchmark VerifiedScore72%Versus best verified row

Best verified: Claude Opus 5 · 96%

Gap24 behindWeightWeighted 16%
Agentic3 rows
Agentic benchmark values, best verified comparison, weight, and source status
Terminal-Bench 2.0Score49.4%Versus best verified row

Best verified: GPT-5.6 Sol · 91.9%

Gap42.5 behindWeightWeighted 38%
OSWorld-VerifiedScore58%Versus best verified row

Best verified: Qwen3.8 Max · 86.1%

Gap28.1 behindWeightWeighted 34%
BrowseCompScore63.8%Versus best verified row

Best verified: GPT-5.6 Sol · 92.2%

Gap28.4 behindWeightWeighted 28%
Reasoning1 row
Reasoning benchmark values, best verified comparison, weight, and source status
LongBench v2Score60.2%Versus best verified row

Best verified: Qwen3.8 Max · 66.3%

Gap6.1 behindWeightWeighted 38%
Knowledge3 rows
Knowledge benchmark values, best verified comparison, weight, and source status
MMLU-ProMassive Multitask Language Understanding ProfessionalScore86.7%Versus best verified row

Best verified: Qwen3.7 Max · 89.6%

Gap2.9 behindWeightWeighted 30%
SuperGPQASuperGPQA: Scaling LLM Evaluation Across 285 Graduate DisciplinesScore67.1%Versus best verified row

Best verified: Qwen 3.6 Max (preview) · 73.9%

Gap6.8 behindWeightWeighted 7%
GPQAGraduate-Level Google-Proof Q&AScore86.6%Versus best verified row

Best verified: Sakana Fugu-Ultra · 95.5%

Gap8.9 behindWeightWeighted 7%
Multilingual1 row
Multilingual benchmark values, best verified comparison, weight, and source status
MMLU-ProXScore82.2%Versus best verified row

Best verified: Qwen3.7 Max · 87%

Gap4.8 behindWeightWeighted 100%
Multimodal5 rows
Multimodal benchmark values, best verified comparison, weight, and source status
CharXivCharXiv ReasoningScore77.2%Versus best verified row

Best verified: Claude Mythos 5 · 93.5%

Gap16.3 behindWeightWeighted 25%
MMMUMassive Multi-discipline Multimodal UnderstandingScore83.9%Versus best verified row

Best verified: Qwen3.6-27B · 82.9%

Gap1 behindWeightDisplay only
MMVUMultimodal Multi-disciplinary Video UnderstandingScore74.7%Versus best verified row

Best verified: Qwen3.8 Max · 82.4%

Gap7.7 behindWeightDisplay only
MathVisionScore86.2%Versus best verified row

Best verified: Qwen3.8 Max · 95.2%

Gap9 behindWeightDisplay only
V*Score93.2%Versus best verified row

Best verified: Kimi K2.6 · 96.9%

Gap3.7 behindWeightDisplay only
Inst. Following1 row
Inst. Following benchmark values, best verified comparison, weight, and source status
IFEvalInstruction-Following EvalScore93.4%Versus best verified row

Best verified: Qwen3.5-27B · 95%

Gap1.6 behindWeightWeighted 35%

Lineage

The sequence follows explicit supersedes links. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.

Base entry

How to read this profile

The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.

Qwen3.5-122B-A10B ranks #65 of 217 on the public leaderboard with a score of 59.47/100. Its source-verified position is #44 of 104.

Qwen3.5-122B-A10B is a open weight model with a 262K context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.

Official exact-value benchmark snapshot from the Qwen/Qwen3.5-122B-A10B-FP8 model card. Native context is 262,144 tokens and the model card notes extensibility up to roughly 1M tokens with RoPE scaling.

15 of 431 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.

Its strongest eligible category is Multilingual at #9, while its lowest eligible position is Coding at #80. a well-rounded choice across a range of tasks.

Radar

Qwen3.5-122B-A10B release history

Full release history

Frequently asked questions

How does Qwen3.5-122B-A10B perform overall in AI benchmarks?

Qwen3.5-122B-A10B ranks #65 out of 217 models on the public BenchAlign leaderboard, with a score of 59.47/100. Its evidence status is Supported, and this profile shows 15 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.

Is Qwen3.5-122B-A10B good for knowledge and understanding?

Qwen3.5-122B-A10B ranks #15 out of 55 eligible models for knowledge and understanding, with a public category score of 77.7/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.

Is Qwen3.5-122B-A10B good for coding and programming?

Qwen3.5-122B-A10B ranks #80 out of 134 eligible models for coding and programming, with a public category score of 47.4/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.

Is Qwen3.5-122B-A10B good for reasoning and logic?

Qwen3.5-122B-A10B has source-displayable benchmark coverage for reasoning and logic, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.

Is Qwen3.5-122B-A10B good for agentic tool use and computer tasks?

Qwen3.5-122B-A10B ranks #63 out of 129 eligible models for agentic tool use and computer tasks, with a public category score of 48.6/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.

Is Qwen3.5-122B-A10B good for multimodal and grounded tasks?

Qwen3.5-122B-A10B has source-displayable benchmark coverage for multimodal and grounded tasks, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.

Is Qwen3.5-122B-A10B good for instruction following?

Qwen3.5-122B-A10B ranks #13 out of 38 eligible models for instruction following, with a public category score of 88.4/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.

Is Qwen3.5-122B-A10B good for multilingual tasks?

Qwen3.5-122B-A10B ranks #9 out of 13 eligible models for multilingual tasks, with a public category score of 36.8/100. That places it in the current top ten for this category. Check the underlying rows before treating the aggregate as a workload guarantee.

Is Qwen3.5-122B-A10B open source?

Qwen3.5-122B-A10B is an open-weight model from Alibaba. Its weights can be downloaded for local or hosted deployment, subject to the published license. Open weight does not automatically mean open source: training data and training code may remain private, and commercial restrictions can still apply.

Does Qwen3.5-122B-A10B have full benchmark coverage on BenchLM?

No. Qwen3.5-122B-A10B currently has 31 source-displayable rows across 431 tracked benchmark slots. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.

What is the context window size of Qwen3.5-122B-A10B?

Qwen3.5-122B-A10B has a reported context window of 262K in the exact-model catalog record. The value stays visible, but the profile marks its source link as unavailable instead of presenting it as directly documented. Maximum output length remains separate because providers often publish a different limit.

Last updated August 13, 2026. Runtime fields remain blank until a sourced snapshot exists.

Watch Qwen3.5-122B-A10B in the weekly brief

Get one weekly email when material rank, price, availability, or benchmark evidence changes are worth revisiting.

Read a sample issue

Join 2,000+ readers.