Model comparison
DeepSeek V4 Flash vs GLM-4.6
DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 41.4 on the Noometry Index.
Last verified . 23 shared benchmarks.
DeepSeekSummary
- They share 23 benchmarks with published results for both. DeepSeek V4 Flash scores higher in 7 categories and GLM-4.6 in 1 category; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4 Flash leads 53.7 to 23.7.
- The biggest single-benchmark swing is CritPt: 16.6% for DeepSeek V4 Flash and 1.1% for GLM-4.6.
- DeepSeek V4 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.60 / $2.20 for GLM-4.6.
- DeepSeek V4 Flash accepts more context: 1M tokens versus 205K.
Side by side
| DeepSeek V4 Flash | GLM-4.6 | |
|---|---|---|
| Provider | DeepSeek | Z.ai (Zhipu) |
| Noometry Index | 53.6 | 41.4 |
| Released | 2026-04-24 | 2025-09-30 |
| Weights | Open | Open |
| Context window | 1M | 205K |
| Max output | 393K | 131K |
| Input $ / M tokens | $0.15 | $0.60 |
| Output $ / M tokens | $0.60 | $2.20 |
| Results tracked | 41 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Flash leads
DeepSeek V4 Flash: 47.9 (#59), GLM-4.6: 40.1 (#148)
| Benchmark | DeepSeek V4 Flash | GLM-4.6 |
|---|---|---|
| LMArena WebDev | 1582 | 1340 |
| SciCode | 49.9% | 38.4% |
| LMArena Coding | 1457 | 1449 |
| ALE-Bench | 1,306 | 340.82 |
| FrontierCode | 18.8% | — |
| SWE-bench Verified (bash only) | — | 55.4% |
| WeirdML | 63% | — |
Agentic & Tool Use Not comparable
DeepSeek V4 Flash: —, GLM-4.6: 32.3 (#66)
| Benchmark | DeepSeek V4 Flash | GLM-4.6 |
|---|---|---|
| Terminal-Bench | — | 24.5% |
| Berkeley Function Calling Leaderboard | — | 72.4% |
Reasoning DeepSeek V4 Flash leads
DeepSeek V4 Flash: 53.7 (#30), GLM-4.6: 23.7 (#172)
| Benchmark | DeepSeek V4 Flash | GLM-4.6 |
|---|---|---|
| Kagi LLM Benchmark | 52.2% | 47.4% |
| CritPt | 16.6% | 1.1% |
| LMArena Hard Prompts | 1444 | 1440 |
| ARC-AGI-2 | 61.4% | — |
| SimpleBench | 61.1% | — |
| NYT Connections (extended) | 89.6% | — |
| ARC-AGI-1 | 89% | — |
| Chess Puzzles | 33% | — |
| Mystery Game Puzzles | 34% | — |
| DTBench | 90.9% | — |
| LMCA | 41.7% | — |
| Epoch Capabilities Index | 154.49 | — |
Math DeepSeek V4 Flash leads
DeepSeek V4 Flash: 60.3 (#37), GLM-4.6: 39.1 (#111)
| Benchmark | DeepSeek V4 Flash | GLM-4.6 |
|---|---|---|
| LMArena Math | 1427 | 1432 |
| FrontierMath (Tiers 1-3) | 57.5% | — |
| FrontierMath Tier 4 | 24.4% | — |
| MathArena Final-Answer Competitions | 76.5% | — |
| OTIS Mock AIME 2024-2025 | 94.4% | — |
| ProofBench | 56% | — |
| FrontierMath (Feb 2025 set) | — | 3.8% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge DeepSeek V4 Flash leads
DeepSeek V4 Flash: 55.4 (#48), GLM-4.6: 40.2 (#124)
| Benchmark | DeepSeek V4 Flash | GLM-4.6 |
|---|---|---|
| LMArena Expert | 1441 | 1431 |
| GPQA Diamond | 91% | — |
| SimpleQA Verified | 33.6% | — |
| Vectara Hallucination Rate | — | 9.5% |
Multilingual Too close to call
DeepSeek V4 Flash: 53.0 (#72), GLM-4.6: 53.5 (#66)
| Benchmark | DeepSeek V4 Flash | GLM-4.6 |
|---|---|---|
| LMArena Non-English | 1420 | 1426 |
| LMArena Chinese | 1468 | 1499 |
| LMArena French | 1439 | 1459 |
| LMArena German | 1418 | 1447 |
| LMArena Japanese | 1406 | 1393 |
| LMArena Korean | 1384 | 1400 |
| LMArena Russian | 1428 | 1419 |
| LMArena Spanish | 1436 | 1436 |
Instruction Following Too close to call
DeepSeek V4 Flash: 74.9 (#81), GLM-4.6: 74.3 (#98)
| Benchmark | DeepSeek V4 Flash | GLM-4.6 |
|---|---|---|
| LMArena Instruction Following | 1421 | 1410 |
Long Context Too close to call
DeepSeek V4 Flash: 43.8 (#85), GLM-4.6: 43.4 (#94)
| Benchmark | DeepSeek V4 Flash | GLM-4.6 |
|---|---|---|
| LMArena Longer Query | 1434 | 1422 |
Writing & Preference DeepSeek V4 Flash leads
DeepSeek V4 Flash: 63.8 (#61), GLM-4.6: 61.1 (#90)
| Benchmark | DeepSeek V4 Flash | GLM-4.6 |
|---|---|---|
| LMArena Text | 1432 | 1440 |
| LMArena Creative Writing | 1403 | 1411 |
| EQ-Bench Creative Writing | 1559 | 1411 |
| LMArena Multi-Turn | 1449 | 1427 |
Frequently asked questions
Is DeepSeek V4 Flash better than GLM-4.6?
DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 41.4 on the Noometry Index.
Which is cheaper, DeepSeek V4 Flash or GLM-4.6?
DeepSeek V4 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; GLM-4.6 lists at $0.60 and $2.20.
Is DeepSeek V4 Flash or GLM-4.6 better for coding?
DeepSeek V4 Flash scores higher on coding benchmarks: 47.9 versus 40.1 in the Noometry coding category.
Which has the bigger context window?
DeepSeek V4 Flash does, with 1M tokens against 205K.
How many benchmarks do DeepSeek V4 Flash and GLM-4.6 share?
23 benchmarks have published results for both models. DeepSeek V4 Flash has 41 scored results on Noometry and GLM-4.6 has 29.