Granite 4.2 3B is the pick; Qwen AgentWorld 35B-A3B gained the most

Pick of the week

A 2 GB file that runs on everything we track

Granite 4.2 3B · IBM · Q4_K_M 2.1 GB

Granite 4.2 3B by IBM takes the ninth pick: the largest popularity gain between the readings of 31 October and 6 November among models that run comfortably on a 24 GB card and are not on cooldown. Its Q4_K_M file measures 2.1 GB, read from IBM's own GGUF repository. At 8K context it needs 3.6 GB and runs easily on every card and Mac we track, the 8 GB cards included; at 32K it needs 6.3 GB, runs on the 8 GB cards and runs easily from 11 GB up.

The pick on your card

How Granite 4.2 3B's memory grows with context

The model accepts up to 131,072 tokens of context. Memory = measured weights + computed context cache + runtime, on an RTX 4090's 21.1 GB usable.

What to run this weekend

The largest and the hottest model that fit each card class

"Runs" means our verdict for the measured Q4_K_M file at 8K context is "Runs easily" or "Runs". Largest = most parameters; hottest = highest popularity score. Check your own card

Departures board

Who gained and who lost this week

The score behind these moves is the one the board shows: a published composite of Hub trending, download growth, 30-day downloads and likes. On the pages it reads as popularity. The full board

Price board

What an hour cost last week

Week median = the median of the week's daily medians, glitch days excluded. "Few offers" marks a median over fewer than five offers. Cloud vs. market = cheapest big-cloud list price ÷ the same day's Vast.ai median. What these numbers are

Chart of the week

B200: fourteen days of hourly prices

The B200 made the week's largest move on the price board, +21 % against the week before.

Daily median of Vast.ai's verified offers, UTC days, glitch days excluded by the same rule as the price board. Gaps are days without a reading.

Price moves

The week's largest day-over-day moves

• 4 Nov · A100 at Clore.ai: +58 % ($0.50 → $0.79)
• 5 Nov · B200 at Vast.ai: +21 % ($8.75 → $10.63)
• 3 Nov · H200 at Vast.ai: +21 % ($5.20 → $6.31)

What the pick costs

Granite 4.2 3B, two ways, per day

For 500,000 tokens a day over 4 active hours, the calculator's "regular" preset:

Newly measured

Files measured this week, and where they fit

• gpt-oss 20B (Q4_K_M, 10.8 GB): fits a 16 GB, 24 GB card comfortably at 8K.
• Gemma 4 12B (Q4_K_M, 6.6 GB): fits a 12 GB, 16 GB, 24 GB card comfortably at 8K.
• Nemotron Nano 9B v2 (Q4_K_M, 6.1 GB): fits a 12 GB, 16 GB, 24 GB card comfortably at 8K.
• Ornith 1.5 9B (Q4_K_M, 5.4 GB): fits a 12 GB, 16 GB, 24 GB card comfortably at 8K.
• Granite 4.2 8B (Q4_K_M, 5 GB): fits a 12 GB, 16 GB, 24 GB card comfortably at 8K.
• LFM2.5 8B-A1B (Q4_K_M, 4.8 GB): fits a 12 GB, 16 GB, 24 GB card comfortably at 8K.

Memory from the measured GGUF file plus the computed context cache; "runs" means our verdict is "Runs easily" or "Runs". Check your own card

First reading

Models scored for the first time

• LFM2.5 24B-A2B (Liquid AI): first scored on 4 Nov, now at 58.
• Ornith 1.5 122B (Ornith AI): first scored on 4 Nov, now at 57.

Rent or buy, one number

3,135 rented hours

At last week's Vast.ai median of $0.51 an hour, the RTX 4090's launch list price of $1,599 (NVIDIA launch MSRP, Oct 2022) equals 3,135 rented hours, ≈ 8 years at 8 h/week. Electricity and resale value are left out on both sides.

Class of the week

A100

Last week's Vast.ai median: $0.84 an hour. Against 30 days ago: +10 %. Since we started reading it 46 days ago: +5 %. Cheapest source at the week's end: Clore.ai at $0.62 (marketplace median). The cheapest big cloud, AWS, listed 3.3× the marketplace median. 34 of 49 tracked open models run comfortably on one at 8K context; the largest comfortable one is gpt-oss 120B.

Maker board

Which labs hold the top-10 slots

Counted from the popularity ranking when this issue was built, Monday 05:45 UTC; the 7-day change is the mean per-model change. The full board

New on the radar

Added this week

• LFM3 1.7B (Liquid AI), added 2 Nov
• Radar Cards claimed: 5 in total. The next number of the RTX 4090 card is #2.

From the tower

What changed on the site

• On cooldown and therefore not eligible as the next pick: Granite 4.2 3B (picked 2026-w45), Gemma 4 E4B (picked 2026-w44), Qwen3 8B (picked 2026-w43), LFM2.5 8B-A1B (picked 2026-w42). The rule