Side-by-side comparison · models
Qwen3.8 27B (free) vs Qwen3.8 Max
What we have checked about each, line by line. Every value shows where it came from and when we last confirmed it.
Models
Qwen3.8 27B (free)
Qwen3.8 27B is an open-weight dense vision-language model from Qwen.
Models
Qwen3.8 Max
Qwen3.8 Max is the flagship model in Alibaba's Qwen3.8 series, the general-availability successor to the Qwen3.8 Max Preview.
Checked facts
Line by line
| Measure | Qwen3.8 27B (free) | Qwen3.8 Max |
|---|---|---|
| Price on akashml | $0.050 per million cached input tokens | Not checked yet |
| Price on akashml | $0.225 per million input tokens | Not checked yet |
| Price on akashml | $1.98 per million output tokens | Not checked yet |
| Price on cerebras | $0.990 per million cached input tokens | Not checked yet |
| Price on cerebras | $0.990 per million input tokens | Not checked yet |
| Price on cerebras | $1.49 per million output tokens | Not checked yet |
| Price on chutes | $0.0039 per million cached input tokens | Not checked yet |
| Price on chutes | $0.039 per million input tokens | Not checked yet |
| Price on chutes | $1.25 per million output tokens | Not checked yet |
| Price on cloudflare | $0.050 per million cached input tokens | Not checked yet |
| Price on cloudflare | $0.450 per million input tokens | Not checked yet |
| Price on cloudflare | $3.20 per million output tokens | Not checked yet |
| Price on coreweave | $0.150 per million cached input tokens | Not checked yet |
| Price on coreweave | $0.400 per million input tokens | Not checked yet |
| Price on coreweave | $3.00 per million output tokens | Not checked yet |
| Price on darkbloom | $0.018 per million cached input tokens | Not checked yet |
| Price on darkbloom | $0.035 per million input tokens | Not checked yet |
| Price on darkbloom | $1.50 per million output tokens | Not checked yet |
| Price on deepinfra | $0.037 per million cached input tokens | Not checked yet |
| Price on deepinfra | $0.150 per million input tokens | Not checked yet |
| Price on deepinfra | $1.88 per million output tokens | Not checked yet |
| Price on dekallm | $0.020 per million cached input tokens | Not checked yet |
| Price on dekallm | $0.040 per million input tokens | Not checked yet |
| Price on dekallm | $2.50 per million output tokens | Not checked yet |
| Price on ionstream | $0.020 per million cached input tokens | Not checked yet |
| Price on ionstream | $0.040 per million input tokens | Not checked yet |
| Price on ionstream | $1.95 per million output tokens | Not checked yet |
| Price on mancer | $0.175 per million cached input tokens | Not checked yet |
| Price on mancer | $0.175 per million input tokens | Not checked yet |
| Price on mancer | $2.50 per million output tokens | Not checked yet |
| Price on modal | $0.020 per million cached input tokens | Not checked yet |
| Price on modal | $0.040 per million input tokens | Not checked yet |
| Price on modal | $2.30 per million output tokens | Not checked yet |
| Price on modelrun | $0.090 per million cached input tokens | Not checked yet |
| Price on modelrun | $0.200 per million input tokens | Not checked yet |
| Price on modelrun | $2.50 per million output tokens | Not checked yet |
| Price on near-ai | $0.018 per million cached input tokens | Not checked yet |
| Price on near-ai | $0.040 per million input tokens | Not checked yet |
| Price on near-ai | $1.35 per million output tokens | Not checked yet |
| Price on nebius | $0.100 per million cached input tokens | Not checked yet |
| Price on nebius | $0.450 per million input tokens | Not checked yet |
| Price on nebius | $3.00 per million output tokens | Not checked yet |
| Price on novita | $0.085 per million cached input tokens | Not checked yet |
| Price on novita | $0.420 per million input tokens | Not checked yet |
| Price on novita | $3.00 per million output tokens | Not checked yet |
| Price on parasail | $0.050 per million cached input tokens | Not checked yet |
| Price on parasail | $0.240 per million input tokens | Not checked yet |
| Price on parasail | $2.20 per million output tokens | Not checked yet |
| Price on phala | $0.050 per million cached input tokens | Not checked yet |
| Price on phala | $0.200 per million input tokens | Not checked yet |
| Latest version | qwen/qwen3.8-27b-20260814 | Not checked yet |
| Time to first token (ms) | 4,060 | 3,180 |
| Time to first token (ms) on akashml | 601 | Not checked yet |
| Time to first token (ms) on cerebras | 248 | Not checked yet |
| Time to first token (ms) on chutes | 2,150 | Not checked yet |
| Time to first token (ms) on cloudflare | 512 | Not checked yet |
| Time to first token (ms) on coreweave | 3,004 | Not checked yet |
| Time to first token (ms) on darkbloom | 1,018 | Not checked yet |
| Time to first token (ms) on deepinfra | 671 | Not checked yet |
| Time to first token (ms) on dekallm | 594 | Not checked yet |
| Time to first token (ms) on ionstream | 887 | Not checked yet |
| Time to first token (ms) on mancer | 722 | Not checked yet |
| Time to first token (ms) on modal | 685 | Not checked yet |
| Time to first token (ms) on modelrun | 334 | Not checked yet |
| Time to first token (ms) on near-ai | 1,856 | Not checked yet |
| Time to first token (ms) on nebius | 655 | Not checked yet |
| Time to first token (ms) on novita | 3,263 | Not checked yet |
| Time to first token (ms) on parasail | 1,087 | Not checked yet |
| Time to first token (ms) on phala | 818 | Not checked yet |
| Time to first token (ms) on prime-intellect | 1,093 | Not checked yet |
| Time to first token (ms) on qwen | 4,153 | Not checked yet |
| Time to first token (ms) on reka | 860 | Not checked yet |
| Time to first token (ms) on streamlake | 5,713 | Not checked yet |
| Time to first token (ms) on tenstorrent | 470 | Not checked yet |
| Time to first token (ms) on venice | 555 | Not checked yet |
| Time to first token (ms) on wafer | 824 | Not checked yet |
| Output speed (tokens a second) | 38.3 | 38.2 |
| Output speed (tokens a second) on akashml | 57 | Not checked yet |
| Output speed (tokens a second) on cerebras | 678 | Not checked yet |
| Output speed (tokens a second) on chutes | 29 | Not checked yet |
| Output speed (tokens a second) on cloudflare | 35 | Not checked yet |
| Output speed (tokens a second) on coreweave | 67 | Not checked yet |
| Output speed (tokens a second) on darkbloom | 92 | Not checked yet |
| Output speed (tokens a second) on deepinfra | 26 | Not checked yet |
| Output speed (tokens a second) on dekallm | 83 | Not checked yet |
| Output speed (tokens a second) on ionstream | 63 | Not checked yet |
| Output speed (tokens a second) on mancer | 64 | Not checked yet |
| Output speed (tokens a second) on modal | 109 | Not checked yet |
| Output speed (tokens a second) on modelrun | 92 | Not checked yet |
| Output speed (tokens a second) on near-ai | 51 | Not checked yet |
| Output speed (tokens a second) on nebius | 149 | Not checked yet |
| Arc agi 2 | 42.4% (for 14 Aug 2026) | Not checked yet |
| Apex agents | 47.5% (for 14 Aug 2026) | 63.3% (for 2 Aug 2026) |
| Arc agi | 87.5% (for 14 Aug 2026) | Not checked yet |
| Dtbench | 88% (for 14 Aug 2026) | 92% (for 2 Aug 2026) |
| Lmca | 41.4% (for 14 Aug 2026) | 46.2% (for 2 Aug 2026) |
| Proofbench | 16% (for 14 Aug 2026) | 58% (for 2 Aug 2026) |
| Surface evolver bench | 45% (for 14 Aug 2026) | Not checked yet |
| Artificial Analysis Intelligence Index | Not checked yet | 53.4 |
| Artificial Analysis Coding Index | Not checked yet | 68.9 |
| Artificial Analysis Agentic Index | Not checked yet | 49.9 |
| LMArena text rating (Elo) | Not checked yet | 1,484 |
| LMArena web development rating (Elo) | Not checked yet | 1,677 |
| LMArena vision rating (Elo) | Not checked yet | 1,313 |
| Design Arena rating (Elo) | Not checked yet | 1,333 |
| Gpqa diamond | Not checked yet | 92.7% (for 2 Aug 2026) |
| Frontiermath tiers 1 3 v2 private | Not checked yet | 74.7% (for 2 Aug 2026) |
| Simpleqa verified | Not checked yet | 45.8% (for 2 Aug 2026) |
| Chess puzzles | Not checked yet | 46% (for 2 Aug 2026) |
| Frontiermath tier 4 v2 private | Not checked yet | 46.3% (for 2 Aug 2026) |
| Frontierswe | Not checked yet | 15.8% (for 2 Aug 2026) |
| Mystery game puzzles | Not checked yet | 38% (for 2 Aug 2026) |
| Otis mock aime 2024 2025 | Not checked yet | 99.4% (saturated benchmark: the top models all score near the ceiling) (for 2 Aug 2026) |
A line appears only when at least one record has a value we confirmed from a source. Values we have not confirmed are left out, never guessed. Records of different types share fewer lines.
Add a third
Compare with one more record
How this works
Share it, or build your own
This page works for any two or three records in the directory, of any type. The link stays the same whichever order you name them in, so you can share it as it is. Change the records in the address, for example /directories/vs/claude-code-vs-cursor, or start from our editorial comparisons.