llmrun

Intel · Alchemist · 16 GB VRAM tier

Best AI Models for Intel Arc A770 16GB

Runs 14B models well; 27–31B only just fits.

16 GB GDDR6560 GB/s131 TFLOPS FP16225 WAlchemist 2022
Buyabout $349

What LLMs can the Arc A770 16GB run?

#ModelSpeedllmrun ScoreQuant · VRAM
1
Qwen3.5 9BAlibabaHow to run →
Q4_K_M · 6.6 GB · 44.0 t/s
44.0 tok/sQ4_K_M · 6.6 GB
2
Qwen3.5 4BAlibaba
Q4_K_M · 3.4 GB · 85.9 t/s
85.9 tok/sQ4_K_M · 3.4 GB
3
GPT OSS 20BOpenAI
Q4_K_M · 13.1 GB · 71.2 t/s
71.2 tok/sQ4_K_M · 13.1 GB
4
Phi 4Microsoft
Q4_K_M · 12.5 GB · 29.4 t/s
29.4 tok/sQ4_K_M · 12.5 GB
5
Gemma 4 12B ITGoogle
Q4_K_M · 9.0 GB · 34.0 t/s
34.0 tok/sQ4_K_M · 9.0 GB
See all 307 models on the Arc A770 16GB →Best quant that fits · speeds estimated on this card

About the Arc A770 16GB

The Arc A770 16GB is an entry card in the 16 GB class: it runs 14B models well. It is power-hungry for its speed, and at its $349 MSRP it gave the best speed per dollar of the cards compared. Alchemist generation, released 2022.

Speed in 16 GB#14560 GB/s
Per watt#110.20 tok/s per W on Qwen3.5 9B (9B params)
Per dollar#112.6 tok/s per $100 on Qwen3.5 9B (9B params)
GenerationAlchemistreleased 2022

Arc A770 16GB strengths and weaknesses

Strengths

  • Runs 35B MoE models: Qwen3.6 35B A3B at 65.0 tok/s in 15.7 GB

Weaknesses

  • 70B models don't fit: Llama 3.3 70B Instruct needs 51.3 GB at Q4 vs 16 GB here
  • oneAPI, not CUDA: No FP8 or NVFP4 weights and no vLLM AWQ or GPTQ; use GGUF builds

Speed per dollar and per watt

On Qwen3.5 9B (9B params) at Q4_K_M. New prices are compared with new, used with used. The last two columns are the speed you get per $100 of price and per 100 W of board power; higher is better. The Arc A770 16GB has no current price, so it is shown at its MSRP.

CardNew pricetok/stok/s per $100tok/s per 100 W
Arc A770 16GB16 GB
about $34944.0
12.6
19.6
RTX 507012 GB
about $84068.7
8.2
27.5
RX 7900 XTX24 GB
about $1,13090.6
8.0
25.5
RTX 5070 Ti16 GBBest new value
about $1,17091.6
7.8
30.5
RX 907016 GB
about $80060.4
7.6
27.5
RX 9070 XT16 GB
about $81560.4
7.4
19.9
RTX 508016 GB
about $1,69098.1
5.8
27.3
RX 9060 XT 16GB16 GB
about $59030.2
5.1
18.9
CardUsed pricetok/stok/s per $100tok/s per 100 W
RTX 3060 12GB12 GBBest used value
about $32936.8
11.2
21.6
RX 907016 GB
about $65060.4
9.3
27.5
RX 7900 XTX24 GB
about $1,00090.6
9.1
25.5
RTX 407012 GB
about $57551.5
9.0
25.8
RTX 5070 Ti16 GB
about $1,10091.6
8.3
30.5
RTX 4070 Ti SUPER16 GB
about $90068.7
7.6
24.1
RTX 4070 SUPER12 GB
about $67551.5
7.6
23.4
RTX 4070 Ti12 GB
about $72251.5
7.1
18.1

Cost per token: the Arc A770 16GB against 12–32 GB cards

On Qwen3.5 9B (9B params, Q4_K_M, 4K context). Speed, cost per token and power per token depend on the model.

12 GB16 GB24 GB32 GBthis cardat MSRP
$0.00$0.50$1.00$1.50$2.00$2.50$0$2,000$4,000$6,000$ per 1M output tokens · lower is betternew price →RX 9070 · 16 GB · ≈$800 · $0.99 per 1M tokensRX 7900 XTX · 24 GB · ≈$1,130 · $0.95 per 1M tokensRTX 5070 Ti · 16 GB · ≈$1,170 · $0.95 per 1M tokensRTX 5090 · 32 GB · ≈$5,376 · $1.99 per 1M tokensRTX 5080 · 16 GB · ≈$1,690 · $1.25 per 1M tokensRTX 5070 · 12 GB · ≈$840 · $0.93 per 1M tokensRX 9070 XT · 16 GB · ≈$815 · $1.07 per 1M tokensRX 9060 XT 16GB · 16 GB · ≈$590 · $1.46 per 1M tokensArc A770 16GB · 16 GB · $349 MSRP · $0.72 per 1M tokensArc A770 16GBRTX 5070 TiRTX 5090RTX 5080RTX 5070RX 9070 XTRX 9060 XT 16GB
$0.00$0.50$1.00$1.50$2.00$2.50$0$2k$4k$6k$ per 1M output tokens · lower is betternew price →RX 9070 · 16 GB · ≈$800 · $0.99 per 1M tokensRX 7900 XTX · 24 GB · ≈$1,130 · $0.95 per 1M tokensRTX 5070 Ti · 16 GB · ≈$1,170 · $0.95 per 1M tokensRTX 5090 · 32 GB · ≈$5,376 · $1.99 per 1M tokensRTX 5080 · 16 GB · ≈$1,690 · $1.25 per 1M tokensRTX 5070 · 12 GB · ≈$840 · $0.93 per 1M tokensRX 9070 XT · 16 GB · ≈$815 · $1.07 per 1M tokensRX 9060 XT 16GB · 16 GB · ≈$590 · $1.46 per 1M tokensArc A770 16GB · 16 GB · $349 MSRP · $0.72 per 1M tokensArc A770 16GBRTX 5070 TiRTX 5090
Cost per 1M output tokenson Qwen3.5 9B · card over 3 years + power
$0.72
Card, MSRP
$0.50
Power
$0.21
3-year cost of ownership$497
Or rent an RTX 3090 Ti in the cloud$0.25/hr

Renting costs as much as buying after ≈1,400 hours (≈1.0 years at 4 h a day).

Include card price

Each dot is a card; hover or tap it for its numbers, click to open it. Cards without a price for the chosen basis are left out. How we calculate cost per token

Similar GPUs

Cards people weigh against the Arc A770 16GB, each with the one reason it is worth a look. Speeds on Qwen3.5 9B.

37% faster, same 16 GBRX 9070AMD · 16 GB · RDNA 4
▲ 37% speedsame 16 GBabout $800
56% faster, same 16 GBRTX 4070 Ti SUPERNVIDIA · 16 GB · Ada Lovelace
▲ 56% speedsame 16 GBabout $799rent $0.15/hr
8 GB more and nearly 2× the speedRX 7900 XTXAMD · 24 GB · RDNA 3
▲ 106% speed▲ 8 GBabout $1,130
4 GB more and 72% fasterRX 7900 XTAMD · 20 GB · RDNA 3
▲ 72% speed▲ 4 GBabout $899
More efficient per wattRTX 5090 Laptop GPUNVIDIA · 24 GB · Blackwell
▲ 108% speed▲ 8 GB
Best value near the Arc A770 16GBRTX 5070 TiNVIDIA · 16 GB · Blackwell
▲ 108% speedsame 16 GBabout $1,170rent $0.16/hr

Arc A770 16GB tokens per second by model size

Output speed depends on memory bandwidth (560 GB/s). Grey rows need more than 16 GB.

  1. 9BQwen3.5 9B · Q4_K_M · 6.6 GB
    44.0tok/s
  2. 14BQwen3 14B · Q4_K_M · 11.9 GB
    29.5tok/s
  3. 27–31BQwen3.8 27B · Q3_K_L · 15.4 GB tight · Q3_K_L
    18.7tok/s
  4. 35B MoEQwen3.6 35B A3B · IQ3_S · 15.7 GB tight · IQ3_S
    65.0tok/s
  5. 70BLlama 3.3 70B Instruct · not at 4-bit
    needs 51.3 GB
  6. 120B MoEGPT OSS 120B · not at 4-bit
    needs 70.8 GB

Estimated speeds: Q4_K_M, 4K context, one user. How we estimate

All 1057 models on the Arc A770 16GB

Featured only

Best quant that fits, ordered by llmrun Score. Fit grades how well it runs here, S best. How ranking works

LLM models compatible with Intel Arc A770 16GB — ranked by Score, recency and Fit
ModelVRAMScoreFit
Gemma 2 9B IT
Q4_K_M·45.9 t/s
7.3 GBA78, Runs well
Qwen3 1.7B
Q4_K_M·250.0 t/s
2.8 GBS100, Runs great
Hy MT2 30B A3B
Q3_K_M·63.4 t/s
15.8 GBC44, Usable
Param 1 2.9B Instruct
BF16·44.5 t/s
7.1 GBA77, Runs well
Datarus R1 14B Preview
Q4_K_M·29.3 t/s
12.4 GBA70, Runs well
OlmOCR 2 7B 1025
Q4_K_M·51.9 t/s
6.2 GBA81, Runs well
Llama 3.2 Korean Bllossom 3B
Q4_K_M·113.8 t/s
4.1 GBS94, Runs great
Qwen2.5 VL 3B Instruct
Q4_K_M·106.5 t/s
3.2 GBS92, Runs great
Granite Docling 258M
Q4_K_M·560.0 t/s
0.6 GBS100, Runs great
Holo4 35B A3B
Q3_K_S·70.2 t/s
15.8 GBC44, Usable
Gemma 3n E4B IT
Q4_K_M·54.1 t/s
7.7 GBA81, Runs well
Sarvam 1
Q4_K_M·136.6 t/s
2.8 GBS96, Runs great
OpenELM 1 1B Instruct
Q4_K_M·394.4 t/s
2.2 GBS100, Runs great
Stories15M MOE
Q4_K_M·465.5 t/s
0.3 GBS100, Runs great
Yi 1.5 9B Chat
Q4_K_M·48.3 t/s
6 GBA79, Runs well

Arc A770 16GB specs, power and PSU

Specs

Architecture
Alchemist
VRAM
16 GB GDDR6
Memory bandwidth
560.0 GB/s
FP16 (dense)
131.00 TFLOPS
TDP
225 W
Released
2022-10-12
MSRP
$349

Arc A770 16GB FAQ

Is the Arc A770 16GB still worth it in 2026?

At its MSRP of $349 you get 12.6 tok/s per $100 on Qwen3.5 9B, which ranks 1st of the 8 cards compared here; it is also the 14th fastest out of 17 cards with 16 GB. It runs 14B models well; 27–31B only just fits.

What is the largest model the Arc A770 16GB can run?

Gemma 4 26B A4B IT (25.8B) is the largest at 4-bit or better: at Q4_K_S it uses 15.5 GB of 16 GB, and runs at about 65.7 tok/s. At 3-bit, a bigger model fits too: Gemma 2 27B IT (27.2B) at IQ3_M, 15.0 GB.

Can the Arc A770 16GB run Qwen3.5 9B?

Yes. Qwen3.5 9B fits at Q4_K_M, using 6.6 GB of 16 GB, and it is the best model that runs well on this card.

Can the Arc A770 16GB run 70B models?

No. Llama 3.3 70B Instruct needs 51.3 GB at Q4, more than the 16 GB on this card.

How many tokens per second does the Arc A770 16GB give?

About 44.0 tok/s on Qwen3.5 9B at Q4_K_M, with the first token after 3.4 s on a 4K-token prompt.

What power supply does the Arc A770 16GB need?

Plan a 450 W PSU (2× the 225 W TDP).

Is 16 GB enough for coding and agents?

Yes for coding: GPT OSS 20B, the best of the 2 coding models that fit, runs at about 71 tok/s. For agents, not yet: no agent-scored model fits in less than 96 GB.

Is renting cheaper than buying an Arc A770 16GB?

An RTX 3090 Ti (24 GB) rents from $0.25/hr. Renting costs as much as buying after ≈1,400 hours (≈1.0 years at 4 h a day).

Related

Best coding LLMs16 GB VRAM24 GB VRAMGPUs with 16 GB or moreRTX 5090RTX 3090 Ti